All 500 of the World's Top 500 Supercomputers Are Running Linux (zdnet.com)

Freshly Exhumed shares a report from ZDnet: Linux rules supercomputing. This day has been coming since 1998, when Linux first appeared on the TOP500 Supercomputer list. Today, it finally happened: All 500 of the world's fastest supercomputers are running Linux. The last two non-Linux systems, a pair of Chinese IBM POWER computers running AIX, dropped off the November 2017 TOP500 Supercomputer list. When the first TOP500 supercomputer list was compiled in June 1993, Linux was barely more than a toy. It hadn't even adopted Tux as its mascot yet. It didn't take long for Linux to start its march on supercomputing.

From when it first appeared on the TOP500 in 1998, Linux was on its way to the top. Before Linux took the lead, Unix was supercomputing's top operating system. Since 2003, the TOP500 was on its way to Linux domination. By 2004, Linux had taken the lead for good. This happened for two reasons: First, since most of the world's top supercomputers are research machines built for specialized tasks, each machine is a standalone project with unique characteristics and optimization requirements. To save costs, no one wants to develop a custom operating system for each of these systems. With Linux, however, research teams can easily modify and optimize Linux's open-source code to their one-off designs. The semiannual TOP500 Supercomputer List was released yesterday. It also shows that China now claims 202 systems within the TOP500, while the United States claims 143 systems.

Comments

That's because...

[Anonymous Coward]

it's made in the USA, USA, USA!

Re: That's because...

[Anonymous Coward]

Actually it's more likely made in China.

Re:That's because...

[Tough Love]

Linux was originally made in Finland.

Re:That's because...

[anon mouse-cow-aard]

oblig: other OS's are Finnished?

Re: That's because...

[Anonymous Coward]

huh, no VMS?

Re:That's because...

[Anonymous Coward]

So was Linus.

Re:That's because...

[Anonymous Coward]

Linus lives in America now, since he abandoned his homeland in search of money, money, money.

Anybody still believe Linus the liar Torvalds about how Linux was just for fun?

Re:That's because...

[Anonymous Coward]

The two are not mutually exclusive. I always chose the more fun (read interesting) job over the one that paid much more (see dotcom). Nevertheless, I was always paid very well. money, money, money

Re:That's because...

[93 Escort Wagon]

Actually he abandoned his homeland in search of warmth.

Re:That's because...

[Anonymous Coward]

Actually, there's a lot of people who did things for fun, that later became a career as they get older and their kids need food... if you ever leave your mom's basement you'll understand.

Re:That's because...

[YukariHirai]

Anybody still believe Linus Torvalds about how Linux was just for fun?

Of course. Linux was just for fun; now he makes a living out of it. A person's motivations for doing something don't have to remain exactly the same for the whole time they do it.

Re:That's because...

[invictusvoyd]

shhh.. No one knows that all the TOP 500 supercomputers run on EMACS..

Re:That's because...

[The Evil Atheist]

If you ask Linus, he'd tell you he still does it for fun.

Re:That's because...

[bruce_the_loon]

After optimizing his computer to the point the CPU couldn't produce enough heat to warm his house anymore.

Re:That's because...

[zifn4b]

it's made in the USA, USA, USA!

No it's because they're running Beowulf Clusters. Sorry couldn't resist and hadn't seen it in a long time. </nostalgia>

Re:That's because...

[zifn4b]

Linus lives in America now, since he abandoned his homeland in search of money, money, money.

Linux 2: The Search for More Money

Linux: The Breakfast Cereal
Linux: The Toilet Paper
Linux: The Flame Thrower (people really love this one and Linus is no stranger to spewing flames)

Re:That's because...

[donaldm]

shhh.. No one knows that all the TOP 500 supercomputers run on EMACS..

No they run under vi but don't tell Richard that. Now let the flame wars of the 1980's begin again. :-)

Re:That's because...

[K. S. Kyosuke]

Actually he abandoned his homeland in search of warmth.

He couldn't have bought a Pentium 4?

Re:That's because...

[barbariccow]

Won't happen. Everyone knows vi is better than emacs.

Re:That's because...

[barbariccow]

I used a newer beowulf cluster just last year.

This is the year

[Anonymous Coward]

Linux makes it to the desktop, of a supercomputer.

Re:This is the year

[MouseR]

Chevy Gen2 Volt's infotainment runs GMLinux. Saw it on the core dump of a crash of another driver's car.

Re:This is the year

[DCFusor]

My 2012 Volt runs linux too. Around the time I bought it, GM was publishing the code for a version of bash they'd done the tighten it up on.

Re: This is the year

[Anonymous Coward]

a crash or a *crash*?

Re: This is the year

[Anonymous Coward]

In this case, would that mean a crash dump is what ends up in your pants if you run off the road?

Re:This is the year

[stooo]

This is the year Linux makes it to the desktop, of all supercomputers.

Re: This is the year

[stooo]

is there Linux in this shit ?

Re:This is the year

[Anonymous Coward]

There is a logical reason for this, and it has nothing to do with Linux.

The Supercomputers level of OSS use is primarily a concern with science. It compiles on multiple platforms, and is well maintained on most of them. Windows and MacOS are only available for the x86-64, ARM, and PPC platforms, and even then, not all of them. That only leaves FreeBSD as an option, and FreeBSD isn't as virtualization friendly, and drivers aren't readily available for GPU systems.

So it's quite literately the only logical choice, owing to that the other choices would have required engineering resources.

That said, Linux does not belong in safety systems, and I hope it never ends up in car automotive systems, power plants, or spacecraft. Everything else is fair game. These systems need real time operating systems that are highly threaded and can respond to events instantly, not be scheduled, or deferred due to eating all the swap space (one of Linux's worst default features, and what makes it woefully awful for web servers by default.)

Re:This is the year

[Tough Love]

...it's quite literately the only logical choice

Oh I know, right? But the big fact you danced around is, Linux is just better than the others. It's faster and more reliable. Otherwise top 500 would not use it. Like, they tried to use Windows, they really did. Microsoft was paying academic institutions to install it and providing teams of free engineers. Still didn't do it. Why? Windows can't handle the load, it can't run continuously under load. It just gets more and more unstable then it falls over. Even when it does stay up, it can't touch the storage, scheduling or memory management efficiency of Linux.

Re:This is the year

[Anonymous Coward]

He didn't mention Windows. And speed is exactly what he was saying Linux isn't the best at. He's referring to varieties of parallel UNIX.

Re:This is the year

[Anonymous Coward]

> The Supercomputers level of OSS use is primarily a concern with science. It compiles on multiple platforms, and is well maintained on most of them. Windows and MacOS are only available for the x86-64, ARM, and PPC platforms, and even then, not all of them.

This makes no sense. Almost all supercomputers are x86-64 based (+/- GPUs).

> That only leaves FreeBSD as an option, and FreeBSD isn't as virtualization friendly, and drivers aren't readily available for GPU systems.

Lol. Supercomputers don't use virtualization.

> So it's quite literately the only logical choice, owing to that the other choices would have required engineering resources.

That's not true, supercomputers within the past 5 years on the top500 list have used Windows, AIX, BSD, Linux. It's just that Linux is better for the job than the others.

> That said, Linux does not belong in safety systems, and I hope it never ends up in car automotive systems, power plants, or spacecraft.

I hope nobody who thinks supercomputers use virtualization ever have their opinion on a computing matter taken by the designer of a safety critical system.

Linux is in safety critical systems already. But it depends on the level and capabilities you're talking about. Processing doppler radar data and sending it to ATC systems in a timely manner is one thing. Running tight control loops in automotive engine and control systems is completely different and just isn't appropriate for Linux.

> Everything else is fair game. These systems need real time operating systems that are highly threaded and can respond to events instantly, not be scheduled, or deferred due to eating all the swap space (one of Linux's worst default features, and what makes it woefully awful for web servers by default.)

You're mixing up all sorts of things here. Nothing responds to interrupts "instantly", what you want is guaranteed hard upper limits. It doesn't even have to be all that fast often times, it just has to be an upper limit so you can design the system to meet response time requirements. Linux can respond "immediately" to interrupts, by the way. It does not have to be "scheduled". Work can be done in interrupt context.

"Highly threaded" what? That's nothing to do with real time.

"Deferred due to eating all swap space" What is this meaningless drivel? Automotive and aircraft control systems don't use swap space. They don't even use virtual memory for god's sake lol.

> (one of Linux's worst default features, and what makes it woefully awful for web servers by default.)

Apparently better than all the others at that too. Windows, OSX, and BSD must *really* be shit if Linux is so bad yet it still beat them all there too.

Re:This is the year

[PeeAitchPee]

The other big logical reason is licensing. No one is ever going to pay for the required Windows or MacOS server licenses for installations of this size when there is free (as in beer) software available. They'd rather put that money into other parts of the project (more cores etc.) to eek out even more performance.

Re:This is the year

[The Cynical Critic]

That said, Linux does not belong in safety systems

Dedicated real time operating systems obviously have their uses, but due to advances in embedded level hardware they're becoming less and less relevant. Even with the overheads of an "almost real time" OS like Linux with some compile switches most modern day embedded hardware is capable of making the dealines in all except some special super low latency use cases. Only place where a real time OS is even necessary these days are rare super low latency and super low power cases (as in under 0.25W).

Serious, 6502s and Z80s are no longer the standard embedded hardware out there anymore.

Re:This is the year

[Xyrus]

It has nothing to do with not being able to handle the load. It has everything to do with costs. Linux is free. Windows isn't. Most of the tools for supercomputing were written for the linux platform. There are tools for windows but since it's a niche market there aren't nearly as many. And since in the super-computing world having a good desktop/GUI environment doesn't mean squat there is no real incentive to use windows outside of certain circumstances.

In all the time I used it I never encountered any serious problems other than the lack of tools and tracking down a couple numerical inconsistencies (issues with compiler differences). It ran fine. It handled loads fine. There's nothing wrong with it's MPI performance, nor any performance issues with it's infiniband stack. If you have a windows admin they can basically hop right on it with little difficulty, and if you're a windows shop then tying it together with AD and such is pretty simple. If you're going full windows, then it might make sense to consider it but there are very few organizations where that's the case.

But there's no getting around the costs.

Re:This is the year

[Junta]

You are right it's not about ability to take on load (though there is a matter of how self-reliant shops can be when trying to analyze failures, which is unlimited with Linux and inherently limited in Windows). However:

the lack of tools and tracking down a couple numerical inconsistencies

Those are pretty huge things. It all stems from the origin of supercomputing as a Unix thing, and as such similarity to Unix allowed seamless porting. Windows however is very different and requires more work to port all of the technical computing ecosystem that no one wants to do (except for a brief period Microsoft themselves, before they figured out just how *much* work they would need to do, how uphill a battle it was philosophically, and how utterly thankless a market win it would be even if they pulled it off).

One could say if the situation were reversed, *maybe* the technical challenges would have been worth solving to get some cost savings, but even if Windows were 100% free, it still wouldn't make inroads into Top500 class systems, because it's a lot of work with no upside to be on Windows.

It's the opposite challenge as Linux on the desktop. Linux on the desktop is perfectly capable and usable now, but there's so much stuff that is Windows only and not enough upside to address all that. We have an entrenched market in both cases and as such you either have to act*exactly* like what you are trying to replace or have to have something *really* worthwhile to convince the market to move.

Re:This is the year

[Ol Olsoc]

It has nothing to do with not being able to handle the load. It has everything to do with costs. Linux is free. Windows isn't.

If I get you right, You spend all this money on a Supercomputer, so you logically use the cheapest OS out there instead of paid ones that should work better?

Sounds legit.

Re:This is the year

[Anonymous Coward]

Actually, it is. The OS, particularly in the unix universe, matters far less than the applications running on it. You think Android uses the Linux kernel because it's "the best?" That's cute.

Re:This is the year

[Anonymous Coward]

Most of the tools for supercomputing were written for the linux platform.

I see reading comprehension isn't your strong point.

Re:This is the year

[Anonymous Coward]

The cost of a Windows license is chump change compared to the cost of an engineering team hacking away on any operating system to get it to run as fast as possible on their hardware.

Re:This is the year

[Anonymous Coward]

You spend lots of money on a supercomputer, therefore you wants the best os for it. In this case, the cheapest os is also the best. No need to pay (a small amount) extra for mediocrity.

Re:This is the year

[barbariccow]

I've used supercomputers that do virtualization. You request a number of CPUs and memory and provide a container or an iso or something and it spits you off a host. It's a pretty sane way to support and delegate 1000s of things that wanna run on it at all times.

Re:This is the year

[Anonymous Coward]

It has nothing to do with not being able to handle the load. It has everything to do with costs. Linux is free. Windows isn't.

If I get you right, You spend all this money on a Supercomputer, so you logically use the cheapest OS out there instead of paid ones that should work better?

Sounds legit.

Right. I was going to respond to the GP in a similar fashion, but I'll add a comment here instead.

If Windows were zero cost, it still wouldn't be used. Yes, cost is a factor ... but openness is far, far more important.

Let's say you spend $<BIGNUM> on a supercomputer. Would you rather:
a) Ask Microsoft to fix some problem you're seeing
b) Ask the community, or fix the problem yourself, or ask RedHat (or SuSE, or ...) .

I suspect getting Microsoft to admit that there was a problem that needed to be fixed would be an ordeal it itself.

Give me open source any day. (Yes, I do hack the kernel when necessary. Fixed an issue last week.)

Further ... even if Microsoft made the OS source available to supercomputer owners, well I have 30 years experience with various UNIX & linux, so I can grok that better. And I'm not alone.

Re:This is the year

[Ol Olsoc]

You spend lots of money on a supercomputer, therefore you wants the best os for it. In this case, the cheapest os is also the best. No need to pay (a small amount) extra for mediocrity.

Exactly

Re:This is the year

[Ol Olsoc]

Most of the tools for supercomputing were written for the linux platform.

I see reading comprehension isn't your strong point.

And responding to the person who actually wrote that isn't yours.

Re:This is the year

[Ol Olsoc]

Actually, it is. The OS, particularly in the unix universe, matters far less than the applications running on it. You think Android uses the Linux kernel because it's "the best?" That's cute.

I think it works, and is consistent.

Re:This is the year

[filesiteguy]

I thought it was QNX. Would love to have OSS in my car. (I have a Volt based Malibu hybrid.)

Re:This is the year

[tbird20d]

That said, Linux does not belong in safety systems, and I hope it never ends up in ... spacecraft.

Too late. Linux is already the Falcon 9 spacecraft.

Re:This is the year

[Anonymous Coward]

That's not virtualization though, that's containerization, or workload management.

We're talking about running computation jobs under an OS which is running as a guest of a hypervisor. I'm not going to say literally none ever have done that because I actually know that has been tried (there have been papers about the practicalities of it). It's just that it is practically unheard of in supercomputers (compute nodes at least, management or IO nodes may be used a bit -- but it is not uncommon to use different OSes on those nodes so the idea that one OS being "not as virtualization friendly" would put it out of contention for running on compute nodes is ridiculous).

Re:This is the year

[Trogre]

Funny, I don't recall my SLES support contracts being free.

Re:This is the year

[Anonymous Coward]

How does this nonsense get modded up?

Microsoft was paying supercomputer builders to use Windows.
Windows is so horrible they couldn't pay the builders enough.

Re:This is the year

[Anonymous Coward]

Actually, both virtualisation and container technologies are indeed in use on supercomputers. (Source: I do this every day on a couple of Top500 machines...)

Re:This is the year

[lsatenstein]

...it's quite literately the only logical choice

Oh I know, right? But the big fact you danced around is, Linux is just better than the others. It's faster and more reliable. Otherwise top 500 would not use it. Like, they tried to use Windows, they really did. Microsoft was paying academic institutions to install it and providing teams of free engineers. Still didn't do it. Why? Windows can't handle the load, it can't run continuously under load. It just gets more and more unstable then it falls over. Even when it does stay up, it can't touch the storage, scheduling or memory management efficiency of Linux.

Microsoft now offers Linux for servers, so I am told. Many of their cloud platforms run on Linux

Re:This is the year

[cthulhu11]

The skewing factor here is that arguably clusters of individual systems really aren't supercomputers, the same way that a string of salps isn't a single animal. With traditional vector / SIMD systems (Cray, Convex) mostly a thing of the past, the term "supercomputer" should be as well, just as we stopped using "minicomputer" when they were no longer part of the equation.

Re:This is the year

[Tough Love]

Microsoft now offers Linux for servers, so I am told. Many of their cloud platforms run on Linux

"If you can't beat them, join them." It's sound business practice, but it does not mean Microsoft's dirty tricks culture is over. The real watershed moment will arrive when they start contributing to interoperability with Libreoffice and Samba because, you know, it makes good business sense to play well with others.

Re:This is the year

[minstrelmike]

You mean you have daemons in your car? Way kewl.

Re:This is the year

[JohnStock]

No you use one you can personally customise specifically for supercomputer applications which Windows was not designed for. Are Linux fanbois that desperate to impress these days?

Re:This is the year

[jon3k]

Windows isn't.

It's damn sure free if you're running it on one of the Top 500 supercomputers in the world, I guarantee it.

Re:This is the year

[Ol Olsoc]

No you use one you can personally customise specifically for supercomputer applications which Windows was not designed for. Are Linux fanbois that desperate to impress these days?

You'd have to ask one, I don't know how desperate they are. I have zero fucks to give about who runs what OS on what.

I just make note that considering the cost of a supercomputer, it makes sense to use an operating system that you consider a good operating system, regardless of the cost. Cost of the OS is way down in the noise in that arena. I use MacOS, UNIX, Linux, and Windows. Whichever works best for what I'm doing.

Your projection of me as a "fanboi" is about as close to a straw man when as you can get. I have my favorite(s), but that is not relevent to this conversation. So chillaxe, have a nice adult beverage, and don't assume what other people are or are not.

'This happened for two reasons.'

[Anonymous Coward]

You only gave me one! I want the second! Oh God, oh God, what's the second reason????

Re:'This happened for two reasons.'

[Anonymous Coward]

There is no second reason.

Linux is not used because it's better, it's used because it is cheaper.

In the end, cheaper almost always wins over better.

Re: 'This happened for two reasons.'

[Anonymous Coward]

Odd statement, considering Microsoft mantra declares Linux is far more expensive than Linux.

Re: 'This happened for two reasons.'

[qwerty shrdlu]

Odd statement, considering Microsoft mantra declares Linux is far more expensive than Linux.

I think you got that backwards

Re: 'This happened for two reasons.'

[repka]

I think he got it just right.

Re: 'This happened for two reasons.'

[PixetaledPikachu]

Only Linux can topple Linux

Re:'This happened for two reasons.'

[jrumney]

I think the cliffhanger ending is the editors attempt to bring back slashdotting. There was once a time, when sites would be brought to their knees by a front page story on Slashdot. These days noone reads TFA, so the concept of slashdotting has faded from memory.

Re:'This happened for two reasons.'

[Anonymous Coward]

Bullshit.

Linux is used because it's far, far, FAR more flexible, less resource intensive and more efficient than Windows, while supporting and making good use of vastly larger amounts of RAM and CPUs.

If you baseline is one of the proprietary Unices, it's still more flexible, less archaic and more familiar to users while supporting a wider range of hardware while being infinitely cheaper.

Re:'This happened for two reasons.'

[Anonymous Coward]

Is it now? Then why did microsoft give away windows lincenses for those who agreed to use windows on ther hpc system. And not only that they even chipped in and subsised the hardware. And still most universites wiped the system after install and press release. One would think that it would be more expensive to run linux in those cases but linux was still the premiere option.

Re:'This happened for two reasons.'

[Megol]

Price is a positive thing of course but not why it is used - the cost for OS software in a supercomputer would be a fraction of the hardware and infrastructure costs anyway.

The thing is that Linux have excellent scalability when it comes to I/O throughput, this is something that many companies and individuals have worked hard to achieve. So it is possible to adapt an OS installation to be suitable for extreme throughput.

The compute nodes themselves doesn't really need a proper operating system (and many supercomputers/clusters in the past had extremely limited systems) and in fact any OS overhead is processing power wasted. Userspace programs using the MPI to communicate is the norm. And again the adaptability of Linux is an advantage, customizing a small efficient kernel with the necessities and nothing else is easy.

But the most important thing is that people already have done the adaptions and that those are available for others to use. Sure there are system specific things that have to be changed anyway but that would be the case for any large cluster machine. The closest to plug and play one can come.

Re: 'This happened for two reasons.'

[bytestorm]

Regarding things that break over time in linux, at least these days, if I static link with musl I can guarantee my binaries will work on other distributions as opposed to being jacked by glibc. Granted nsswitch will be broken if you want to use anything but dns, but screw it!

Re: 'This happened for two reasons.'

[Anonymous Coward]

In a Microsoft "Office" environment, you pay licencing costs, but the support and training is provided.

By whom? Sorry, but in an MS environment support is paid for by the company. Either you pay MS to support it or you pay trained techs to support it. And what happens WHEN MS decides to shuffle Windows around? The company pays to retrain all those techs they hired or they renegotiate with MS to support the system for them -- AGAIN. Where do you people get this "support is free" shit?

You can not plop down a linux desktop in front of someone and have them use it.

Who can't use a default linux desktop like Ubuntu or Mint? Every time my company upgrades the desktop OS to the newest version of Windows I need to sit down with all the users and show them where their little icons are and how to access the features they never use and how to log in and log out. I do the same thing with Linux users. This is just the way the world works.

Linux's inability to not break itself over time is another major cost source. Windows may break itself sometimes, but hasn't been so shitty since Windows ME.

You are absolutely right! Why, when I installed Steam on WINE, I totally hosed my entire....

Hold it. Plain vanilla users don't do things like that. They sit in front of their machines and enter data into interfaces, create spreadsheets, write documents and work with email. All of that stuff works fine. Rather, I should say all of that stuff works just as well as it does on Windows. Case in point: some joker developed an app in Access 2000 about sixteen years ago and one of my sites will NOT stop using it and will not upgrade it. It breaks all the time because of dependencies. There is no damned difference between Windows and Linux.

The real fact is that MS hires a metric shit-ton of people to go on forums and spout moronic crap like you just did, so everything you just wrote is suspect.

Re: 'This happened for two reasons.'

[barbariccow]

The hell are you talking about? Maybe static compilation isn't the way to go. You can ship the .o files and link them via an "install.sh" script on the target system, then you can work against multiple versions of libs. Or just.. you know... provide the source.

Doesn't guarantee success on the desktop

[Anonymous Coward]

Unix never made inroads on the desktop.
 
  This might actually be harmful if people think Linux is complicated or designed for heavy hardware they may not consider it for desktops and use cases involving desktop apps.
 
  Linux has been ready for the desktop since about 1999, before that there were dependency issues and hardware wasn't always supported. Now hardware is more likely to be better supported on Linux than on Windows. I'm writing this on Windows but that's only because Windows came on this machine, I'll be installing Linux when I have a week of downtime.
 
  Enlightenment is probably the best looking desktop software anywhere, it's customizability makes it hard to include with distros but it should be considered as evidence that it's not user-friendliness or beauty holding Linux back.
 
  I think it's a bit sad to see Linux software becoming overly simplified in the wake of Apple's success the way other software is.
 
  Linux needs to remain the enthusiast and expert operating system more than it needs broad acceptance. Look what happened with the internet, Linux is great without ads, malware and other problems I associate with popularity.
 
  That said Linux skills are still hugely undervalued and not taught in schools which needs to change. A Linux machine is still your best bet that your machine will still be runnning with data and apps updated but not broken after 10-15 years.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

"I'll be installing Linux when I have a week of downtime."

And that is precisely why Linux isn't a good desktop OS.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

Well said. I was thinking the same thing.

Re:Doesn't guarantee success on the desktop

[c]

Linux has been ready for the desktop since about 1999...

Eh, no, not really. You're talking about a KDE 1.0, pre-Gnome desktop... I used it, but I wouldn't have inflicted it on anyone I needed to support. Five years later it was certainly reasonable, at least where the average non-technical user was concerned.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

I can see that a DE for novice users was not around until the late 1990s, but I had X and Motif under Linux running back in 1993 on a 386. Ran like shit, but it ran. The next machine ,a Pentium, did OK with fvwm.

For an environment that was "good enough", i.e. about on par with CDE, it was fairly easy to accomplish. However, 1995 was a banner year in UI design because of the concept of a Start menu, which everyone cloned except Apple.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

What happens when you click on the apple in the corner of a 1995 MacOS system?

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

What happens when you click on the apple in the corner of a 1995 MacOS system?

i did that and it made me gay

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

Clearly you confused Linux with OSX

Re:Doesn't guarantee success on the desktop

[Kjella]

Eh, no, not really. You're talking about a KDE 1.0, pre-Gnome desktop... I used it, but I wouldn't have inflicted it on anyone I needed to support. Five years later it was certainly reasonable, at least where the average non-technical user was concerned.

KDE was '96, GNOME '97.. in 1999 you'd already have KDE 2.0, didn't use that but I remember trying RHL 6.2 that came out in April 2000 which looks pretty much like a normal desktop to me. Remember that it was going head to head with Windows ME as the consumer desktop, using either was a major PITA. Granted, XP was a big step up but then you had Vista... you can make a lot of excuses for YotLD not happening but that Microsoft brought their A-game is not one of them.

The cornerstone for Microsoft's dominance is Office and Excel in particular, all those people who had to use Windows at work of course took what little knowledge and training they had and bought a Windows machine for home too. When Outlook kicked Lotus Notes to the curb they locked that market up good.

Re:Doesn't guarantee success on the desktop

[ceoyoyo]

Linux still isn't ready for any desktop it isn't installed on. It IS installed on lots of desktops in places like research labs, mine included. But if it's going to make it to anybody else's desk it needs some basic things fixed. I don't know if it's possible to do something as simple as configure a graphics driver in Ubuntu's GUI, but it's certainly not easy.

Everything else works perfectly fine, but none of the GUI systems seem to offer a user friendly way for command line averse users to fiddle with their system settings.

Re:Doesn't guarantee success on the desktop

[argumentsockpuppet]

On a scale of LFS to Mandrake, how bad was your experience?

I'll be installing Linux when I have a week of downtime.

It takes me a couple months to transition a new workstation to Windows. Each time, I try to learn what the native software options are and whether they can meet my current needs. Where it doesn't, I install or use recommended software in most cases to see if it does what I need (WSL) though I do have exceptions for my personal favorite software in a few instances where I'm just unwilling to learn something new. (EditPlus, GIMP, VLC, Sysinternals, Putty.)

A new install of Linux takes the same process, but it has apt or yum or whatever which speeds things up pretty dramatically. With a new Linux desktop install, I just rarely have to learn too many new things... usually. (Eyeballing you hard here systemd!)

If I have to support any significant sized network, and if it's possible, I'd do Linux desktops everywhere. I'd rather use those admin systems than admin Windows... but I work in a job where I have to support Windows because that's all the core software runs on. As an admin, I can do about anything I need to on Microsoft servers and workstations. It'd be false modesty to say I'm not good at admin on Microsoft system. On the other hand, I have used Linux and various BSDs at home and work (on servers) since the late 90's. I could eliminate Microsoft in our workplace and cut our IT departmental work by maybe 30% if only our primary system ran on Linux. I'd miss some of the AD/DHCP/DNS/DFS stack. I'd miss Excel (running native) and Exchange/Outlook, but honestly, running the alternatives in the cloud or Libre would probably reduce our helpdesk workload after a year or two.

I'm good at my jobs, and whatever systems I admin, I'll learn to be good at. Given the ideal scenario, I could run several thousand workstations with the same effort I'd use for a couple hundred Windows workstations. The scenarios I have been hired to handle haven't been ideal, so I've learned to take advantage of the environments I'm in. I'm good at my jobs because I like to learn. I like tinkering, trying new things, scripting and writing real code. That makes me useful, maybe even it helps toward making me valuable.

On the other hand, my varied experiences and experimenting have made me aware that my own weakness is a desire to try new things. If I were designing the systems for a company responsible for my income, it wouldn't be Linux or Windows or Mac. It'd be PC-BSD on the workstations and AIX on the servers. They're boring. Boring is what I look for in a business network. Ideally, the network will be so stable that IT doesn't spend any time working on the backend systems, and that means boring is the goal. I like Linux because I'm always learning new stuff and I like Windows.. sorta, because I'm always being forced to learn new things. That's why I'm sorry to see AIX take a dip off the top 500, but I can see it; Linux is fun.

Re:Doesn't guarantee success on the desktop

[Dutch Gun]

I've sort of made my peace with the fact that desktop Linux will always be a niche player for enthusiasts and experts, plus a tiny percentage of normal people who have experts to admin their systems. There are a lot of reasons for this, but I think that an OS which is still a CLI focused system at its heart has some disadvantages as a desktop solution for the masses.

On the other hand, I think Linux is really in its element with specialized environments like supercomputing. You can hack everything to make the OS work exactly as you need for the hardware you use, and strip out everything that's superfluous. There are lots of great open source tools to take advantage of. There are advantages in drawing from the wide range of scientific talent who are already familiar with using and programming for Linux. And of course, you don't have per-core or per-processor licensing fees for your OS, which is a nice bonus.

Re:Doesn't guarantee success on the desktop

[PixetaledPikachu]

Linux still isn't ready for any desktop it isn't installed on. It IS installed on lots of desktops in places like research labs, mine included. But if it's going to make it to anybody else's desk it needs some basic things fixed. I don't know if it's possible to do something as simple as configure a graphics driver in Ubuntu's GUI, but it's certainly not easy.

Everything else works perfectly fine, but none of the GUI systems seem to offer a user friendly way for command line averse users to fiddle with their system settings.

Both AMD and Nvidia have their config UI packaged to their binary blob driver on linux.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

I don't get why it's always seen as a bad thing that the configuration is in plain text files. I prefer editing a text file to hunting through gpedit or regedit.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

How the hell would you know? Ever tried a new software before? Everything has a learning curve. OS change is much more complicated, because you need to replace programs and actions you already know.

Try switching to OSX (or Windows, if you already are using OSX primarily), it's different and i bet it'll take you more than a week to get used to it and find alternative programs

You are nothing but a Windows' (or OSX's) bitch and your comment could have not been more idiotic.

Re:Doesn't guarantee success on the desktop

[Anne Thwacks]

I don't get why it's always seen as a bad thing that the configuration is in plain text files.

Because if your design is a pile of manure, it is a lot more obvious in plain text files.

(The same reason that most graphics drivers are blobs).

Re: Doesn't guarantee success on the desktop

[stooo]

>> "I'll be installing Linux when I have a week of downtime."
If you need a week of downtime from MS to convince you to switch to Linux, you should rather stay with MS until having a month long downtime. Then you'll be really convinced :)

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

LOL. You've never had to reinstall your Windows machine, I take it? Reinstalling Windows and all it's applications, drivers etc. is a FAR more tedious task than installing Linux.

Linux is an excellent desktop OS. The problem is that most people don't want it, they want Windows because that's what they have been taught, what they know, and comparatively rarely, runs some application which they actually need but can't find a replacement for. (More frequently it's won't find a replacement for.)

None of that is any evidence of Windows being superior, it's just the normal lazy people who either refuse or are too coward to leave their comfort zone. If people were brought up on Linux, the same factors would work in the other direction.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

> Well said. I was thinking the same thing.

Well, two idiots - you have a friend in each other!

Re: Doesn't guarantee success on the desktop

[Interfacer]

If you need a week downtime to install an OS which you already know -which the GP implies-, something is not right.

Re: Doesn't guarantee success on the desktop

[Interfacer]

I've done it several times when building a new machine. You pop in the windows disk and when it is done installing, you install your drivers or apps. The last couple of years I haven't had to use the command line for anything. Depending on the amount of apps and the data you want to move, it takes time but generally not a lot of hassling.

I admit it's been 10 years since I worked with linux. Back in those days, I could install linux fairly easily, but there was always something that required a serious amount of dicking around with config files to get it working. The 3 most common issues would be that sound wouldn't work, wireless wouldn't work, or the graphics config of X (resolution, refresh rates) would be borked.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

pffff. Dude, you need time to backup your data. You need time to categorize the info you have in there. You need time to catalog the software you have, and look for the equivalent and/or alternatives for each of them. And so on.

Does not matter which OS you come from and what OS the destination is. If you change, you need to invest.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

Thank you for an objective response, however their market dominance is not solely due to office. They did not do a bad job with the rest of the system either. NT4.0 was tight. No one will deny that NT4.0 was not more than sufficient as an operating system. The consumer hell that has been created might have changed that landscape, but had they kept that line going they would be the definitive answer for the consumer and workspace operating system. It's true. They had it locked up tight. It was only in the fuck up that only microsoft could accomplish that linux really took off.

Re:Doesn't guarantee success on the desktop

[dmpot]

Unix never made inroads on the desktop.

Commercial Unix workstations had a decent desktop for its time, but they were too expensive. Their private hardware was the primary reason of their excessive cost. Still they were not eager to offer their OS for PC, when PC became power enough to run UNIX OSes. Though, there were some commercial UNIX OS for i386 in the early 1990s, they were too expensive for private users. By his own admission, Linus would never write his own OS if he could just buy a commercial Unix OS for a reasonable price.

Linux has been ready for the desktop since about 1999

I use Linux desktop since 1997, and I know people who started to use it before me. However, when it comes to regular people, Ubuntu became the first distro usable by non-technical crowd. (BTW, personally, I still prefer Debian over Ubuntu, but that's another story...)

I'm writing this on Windows but that's only because Windows came on this machine, I'll be installing Linux when I have a week of downtime.

In this regard, my experience is very different than yours. If I had a machine with pre-installed Windows, it would take me a week to install all software that I regularly use, and then I had to change a lot of default settings, which is really annoying. On the other hand, it takes me one day to install a Linux distro from scratch with all required software, and I am ready to go.

In any case, the reason why we don't have more Linux desktops has nothing to do with the Linux kernel. If you want to have a good desktop, having a good kernel does not really matter that much. On the other hand, if you want to build a supercomputer, the kernel is the crucial part of your system.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

Try switching to OSX (or Windows, if you already are using OSX primarily), it's different and i bet it'll take you more than a week to get used to it and find alternative programs

I have tried switching to osx and windows and really struggled:
- I have to manage windows manually
- Stuff pops up on a random screen
- I keep needing to use the mouse

Normally, if I'm forced to use one of those things, I quickly install putty and ssh to somewhere I can run emacs and a screen session.

OSX and windows are clearly deficient. Anyone who disagrees is also clearly deficient, lacking in education and at best stupid.

Caveat: xmonad and emacs keybindings burned into muscle memory.

Re: Doesn't guarantee success on the desktop

[Interfacer]

Note that I didn't say it was linux's fault. I don't really care whose fault it is / was. All I care about is that when I pop in the installation cd, things get installed and to a decent working state without a lot of finagling. Linux used to have that problem. And that might have various reasons but ultimately, the 'why' doesn't matter to the user.

Honest question: Let's say I take a 2 year old laptop. I know for sure that I can install windows, and that after installation, I will have sound, wifi and a decent screen. Is the same true for linux if I grab a commonly available distro?

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

Of course it matters, because what you are attributing to Windows, "things just working" isn't an effect of Windows, it's a consequence of every bloody OEM and their mother bending over backwards to accommodate it. It has nothing to do with the OS itself, it's a clear sign of incompetence and ignorance to say it does.

That said, your question is framed completely the wrong way. Your random laptop was made to be compatible with Windows. Of course it's going to work. It wasn't made to be compatible with Linux, in some cases (locked bootloader etc) it was specifically made to not be compatible with Linux. Not Linux's fault, yours for buying crap.

However, if you do your research, it's not (yet) exactly hard to find a laptop which is fully functional with Linux out of the box. You just have to exercise your power as a consumer and not buy crap which is incompatible or non-functional. A side note on this is that stuff, notoriously certain WiFi-chips, which do not have support under Linux, are in that state for a reason - usually because they are shit! They don't even work properly under Windows!

On a final note, it's worth pointing out that Windows only works as long on your laptop as Microsoft wants it to. This goes double for laptops with locked bootloaders which are dead the moment Microsoft decides they're not worth releasing updates for. And while Debian hardly is the distro I'd recommend for newbies, it still keeps my AD 2003 Thinkpad working for the tasks I decide it's still capable of.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

I think I broke the sarcasm-o-meter belonging to someone with modpoints. Weak sauce, whimp ass Americans. *shrug*

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

Actually, I bought a E31-80 (or something like that) laptop recently.
The WLAN chip (and everything else) works flawless on Linux.
On Windows, it does not work reliably after a reboot (even Windows to Windows, originally I thought it only happened Linux to Windows, which would have been kind of excusable), I have to power off the computer to make it work again.
And this is an Intel WLAN that (at least used to) have some of the better drivers.
So it actually does happen that on hardware designed and tested to work with Windows, the Windows drivers are still the buggy ones.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

It happens. I have the sneaking suspicion that people writing drivers for Linux usually have a greater need for them to actually be good rather than shippable compared to their Windows counterparts.

Re: Doesn't guarantee success on the desktop

[Zero__Kelvin]

Maybe he isn't an idiot and wants to make sure he has time to handle any unanticipated issues. Migrating your working environment isn't just installing Linux, which can be done in an hour or three. A complete backup of the system needs to be done, and you need time to verify all the data has been restored and you can still do all the things you need to get done. Then, unlike with Windows, you have the ability to customize and tweak to your hearts content.

Re: Doesn't guarantee success on the desktop

[donaldm]

"I'll be installing Linux when I have a week of downtime."

And that is precisely why Linux isn't a good desktop OS.

Time to upgrade from Fedora 26 to Fedora 27 for me was about an hour and I actually do an install upgrade since that gets rid of any rubbish that I have not documented. People document their important desktop requirements - don't they?

Of course, I could do an update process which takes less time and is easier but since I do install some programmes for testing purposes a fresh install does get rid of them unless I have documented that I wish to keep them.

If you take more than three hours to do an upgrade and most of the time you are waiting then you are doing something wrong. Even a fresh install should take no more than one hour (normally about 20 minutes if you use the automatic install and that includes distributions like Mint as well).

So I spend about one to one and half hours every half year on an upgrade is that a problem? I do backup but rarely have to go through a recovery since I don't reformat my data filesystems when doing an install upgrade although I do reformat all system filesystems.

It should be noted that even the stable version of Fedora distribution is not really for the novice since you will be basically on the dull bleeding edge but that said it is still a very good distribution. I would recommend waiting at least a week before installing/upgrading to the latest version unless you are confident in your abilities, although for the last six releases I have not had any issues and issues prior to that have normally been minor.

Re:Doesn't guarantee success on the desktop

[Junta]

Enlightenment is probably the best looking desktop software anywhere, it's customizability makes it hard to include with distros but it should be considered as evidence that it's not user-friendliness or beauty holding Linux back.

Note that 'beauty' is relative and certainly it is not equivalent to 'user friendly'.

I will though agree with the sentiment that there is no winning the 'user friendly', because the main desktop environments are user friendly enough, but there just isn't enough upside for the casual user to bother to even think about changing. As such diminishing the 'enthusiast' experience for the sake of the casual user is a strange thing to do.

Re:Doesn't guarantee success on the desktop

[Dr. Evil]

Outlook didn't kick Lotus Notes to the curb... IBM kicked Lotus Notes to the curb.

Re: Doesn't guarantee success on the desktop

[Interfacer]

Actually, if you want to be prepared and have a solid rollback strategy, you'll remove the HD, insert a fresh one and start. And should something fail that you cannot resolve in a timely manner, you put the old one back. That is how we do the very risky server migrations / software updates at work during the annual downtime window.

As for the backup: you should always have a decent up to date backup. If that is something you need to plan a long time in advance, you're already one hard disk failure away from catastrophe.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

Both AMD and Nvidia have their config UI packaged to their binary blob driver on linux.

So what? That's like saying there's 2 people not burning alive in hell. Were you a Trump voter, per chance?

Re: Doesn't guarantee success on the desktop

[Interfacer]

I do understand what you mean. I am a software developer myself. Since '98 I have developed software for both linux and windows. Since 2007 windows exclusively. So I understand the OEM story and the rest. In fact for a time being I was a very happy BeOS user until that went down the tubes. For a while I participated to OpenBeOS but that went nowhere.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

It takes me a couple months to transition a new workstation to Windows.

This gross level of incompetence would get you fired where I work.

Re: Doesn't guarantee success on the desktop

[donaldm]

Laptops can be problematic with Linux distributions and it is always a good idea to do a search to confirm the Linux distribution will work. I personally have found HP laptops do work although I would still recommend a check. A base desktop (check your graphics card compatibility though) will normally run Linux but it does not hurt to check.

It should go without saying but please image or backup your laptop first. I strongly suggest replacing the disk if you are unsure and even if you get a cheap 120GB SSD you will find that your operating system will take up less then 8GB of data (allocate 30GB to 40GB though) and that will allow you about 80GB to 90GB of user filesystem. Even a 240GB SSD is not that much more expensive and you will get blazing speeds.

Once you have confirmed your particular laptop will run a Linux distribution get a LIVE version on a USB stick or CD-ROM and boot off it. If it boots and brings up the particular window (ie. Gnome, KDE, Xfce, ... etc) manager and run any tests that you require. Only after you are confident that all tests work then you can install.

Installation (assuming no dual boot which I don't recommend) of most Linux distributions especially if you just accept the defaults should take less then twenty minutes.

My Desktop supports 7.1 sound although I only have a 2.1 sound system which if I turn it up will annoy my wife and neighbors. My motherboard supports up to 4K and I have tested it with a 1440p monitor without any issues.

If you are really into the latest games then please ignore what I have said and stick to MS Windows. There are over 5,000 games (see Steam) that will run natively under Linux but they are usually not the latest games. Anyway, I have a PS4 and a backward compatibility PS3 for games so that is not an issue for me.

I run Fedora (latest is 27) however I don't recommend that distribution for novice Linux users although I have not had any issues with it for over three years (Fedora 22) and even then the issues I had were minor. I even have Fedora 26 on an eleven year old HP laptop although it chugs when playing H265 video (H264 is fine). A good distribution for the novice upwards is Mint. If you have a problem or question check out the appropriate forums they are all very good.

Re:Doesn't guarantee success on the desktop

[houghi]

The ONLY think that Linux needs is pre-installed. People do not care what OS they run.They happily use MacOS on the Iphone, Android on their tablet and Windows on their PC or any other combination they can buy.

Those people that do their own install already run what they want and go out of their way to run it.

Re: Doesn't guarantee success on the desktop

[JesseMcDonald]

Honest question: Let's say I take a 2 year old laptop. I know for sure that I can install windows, and that after installation, I will have sound, wifi and a decent screen. Is the same true for linux if I grab a commonly available distro?

Anecdotally... I just ran a Debian live image (from a USB stick) on a roughly two year old laptop to image the hard drive and begin diagnosing an intermittent loss-of-communication issue with the WiFi. The live image immediately started up to a GUI desktop with the proper graphics drivers and perfectly functional WiFi. This was not one of the "user-friendly" desktop distros, and the laptop was not selected for its Linux compatibility, yet—with zero time spent configuring it for that machine—Linux worked better than the Windows environment installed by the manufacturer. (The WiFi issue was eventually fixed by manually deleting and reinstalling the drivers.)

Now, there are systems out there with unsupported WiFi, unsupported graphics chips, and/or firmware that will refuse to boot anything without Microsoft's seal of approval, so I would recommend investigating a bit before making a purchase to ensure there aren't any major Linux incompatibilities. However, these are the exception rather than the rule. Most systems will work out of the box, especially if they have Intel or AMD graphics chips. (Systems with nVidia graphics may require additional closed-source drivers for full acceleration support, depending on the distro.)

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

> What happens when you click on the apple in the corner of a 1995 MacOS system?

You sit there for about 20 minutes while it attempts to copy a 17 Meg file from one folder on the hard drive to another folder?

Re: Doesn't guarantee success on the desktop

[Zero__Kelvin]

Try that approach with a modern laptop. Hint: Swapping the hard drive often involves soldering or otherwise taking the laptop apart these days.

Re:Doesn't guarantee success on the desktop

[c]

KDE was '96, GNOME '97.. in 1999 you'd already have KDE 2.0

KDE 2.0 wasn't released until 2000. And prior to 3.0, KDE was still a bit too rough an average user (I tried, trust me). GNOME didn't have a working release until 1999, for some definition of "working" that I also wouldn't inflict on anyone I needed to support.

I think RHL 9 was really when things came together in a state where a Linux Desktop wasn't an uphill battle.

When Outlook kicked Lotus Notes to the curb they locked that market up good.

I can't give Outlook much credit for that. Lotus Notes was its own worst enemy.

Re: Doesn't guarantee success on the desktop

[Anonymous Coward]

I would like to apologize if I've come off as overly negative, but it seems these days most people who ask if they can do X regarding Linux actually go out of their way to fail in order to be able to claim "it's not ready" or "see, it doesn't work".

Given your explanation I'm sure you understand that especially laptops are potentially hostile territory. As I said, I've never had any problems with it, though I've mainly used machines from HP, ASUS and IBM (yeah, pre-Lenovo). That said, it's impossible due to aforementioned reasons to guarantee anything. Generally speaking though, Linux is ridiculously compatible with all kinds of hardware, but fundamentally it's a hostile world with non-zero number of lemons and land-mines, especially at the cheap end. You have to be careful and do your research. That said, you're probably still at a stage where you're unlucky if you find something which doesn't work, rather than the other way around. Provided that you avoid the systems which are deliberately sabotaged from the OEM, that is.

Re:Doesn't guarantee success on the desktop

[ceoyoyo]

Exactly my point. There are so many different, potentially incompatible, and UI discordant ways to adjust things. That's fine, but what's lacking is a unified system. Some of the distros have tried, but they've mostly failed.

The windows control panel (or whatever they call it these days) is a bit of a UI nightmare, but you can configure whatever a regular user is likely to want from there. Ditto with OS X's Preferences. Linux has everything except a nice, consistent, graphical, go-to, built-in place for the point and click crowd to fiddle with settings.

Re:Doesn't guarantee success on the desktop

[ceoyoyo]

I disagree, from experience. If you set up someone's computer with Linux they don't notice, or care, but they get frustrated when they want to change their desktop image, colour scheme, screensaver, mouse tracking speed or some other trivial setting and can't figure out what icon to click to do it.

Re:Doesn't guarantee success on the desktop

[Anonymous Coward]

I haven't used (open)SuSE for years. But back in the day when I used it, YaST (I think it's called YaST2 these days?) was the best control panel in Linux desktop. I think it still is, if YaST hasn't changed drastically.

Re: Doesn't guarantee success on the desktop

[mattventura]

Not necessarily. The data can just stay in whatever structure it’s in. As for software, you can always start with virtualization and gradually look for alternatives.

Re:Doesn't guarantee success on the desktop

[argumentsockpuppet]

What do you think the PCs we buy come with? Seriously.. "just lol"

The AC comments are strong in this thread. I shouldn't let it bother me, but I wonder how they/you keep track? I can't be bothered to look up my AC comments even the same day, let alone keeping up on a thread like this. I'm genuinely curious what techniques you use? Don't let that curb your trolling, just toss in a tip when you talk about my mom or whatever.

Most of my workstations have started out as Windows, or in one exception a Mac. On the Mac I used Windows with VMWare, on the windows workstations, they usually started out dual booting. My first personal workstation was Windows 3.1, which I used to get started in programming. Later I used it to learn about Linux and BSD. I ran it side by side with a Solaris box for a long time until an unfortunate incident with a screwdriver and escaping magic smoke. (My quite young daughter thought it was awesome Dad was setting off fireworks in the living room.)

I always try to get the new versions of Windows at home before they hit RTM and I need to support them at work. Linux is a little more stable; it's rare (bleh again to systemd) to need to learn much new. I recall fondly installing a new CPU and more RAM in my Windows 3.1 PC so I could install Windows 95. I learned about jumpers and motherboard diagrams for that. I'd say that was when I fell in love with computers as a career. I recall installing Red Hat on that same machine later just to learn about alternative OS designs. I'd say that's when I fell in love with computers as a hobby. I was very fortunate to learn enough to incorporate Linux and Unix into my career as well.

Re:Doesn't guarantee success on the desktop

[PixetaledPikachu]

Exactly my point. There are so many different, potentially incompatible, and UI discordant ways to adjust things. That's fine, but what's lacking is a unified system. Some of the distros have tried, but they've mostly failed.

The windows control panel (or whatever they call it these days) is a bit of a UI nightmare, but you can configure whatever a regular user is likely to want from there. Ditto with OS X's Preferences. Linux has everything except a nice, consistent, graphical, go-to, built-in place for the point and click crowd to fiddle with settings.

In Ubuntu it is managed under a single UI called "settings". Of course Nvidia and AMD have their own config UI, which is also the case with Windows anyway. I recently moved the wife's notebook to Ubuntu and teach her that she can find applications/functions that she needs by typing keywords after pressing the super button. No complain so far. She has managed to change wallpaper, importing files from usb sticks and her android phone, do light image editing without asking me on how to do it

Re:Doesn't guarantee success on the desktop

[argumentsockpuppet]

Wow. That kind of comment is a little too familiar. You call learning new skills, constantly updating your understanding of new options incompetence? In my department, if you're not constantly refining your skills, constantly learning more about the systems you use and support and the alternatives available... well that's what we'd call incompetence. Your comment reminds me of two ex-coworkers who I suspect would have shared your perspective. I admit my reactions may be biased as a result, my own viewpoint is likely colored by that designation: ex-coworkers.

I can scarcely imagine what would have happened if I'd just stuck to the things they taught me in college, instead of poking at every corner of each new release or potential configuration of Windows. Showing my age a bit here, but it'd be hard to find a job where my expertise, even just managing my own workstation, is twenty years out of date. I honestly appreciate the reminder though. I do need to get back in the Windows Insiders Fast Ring. I dropped out for stability issues that actually did manage to irritate me, but I do feel like I need to get back in again, just to keep that one step ahead. The BSODs may have raised my blood pressure and cost me a handful of minutes, but I need the occasional reminder that it's important to focus on the learning instead of the irritants.

All that said, it really doesn't matter what Windows version, or even what OS is on my workstation, since I can jump to any of a half dozen various alternatives. (I worked for years with one of my workstations running Linux and never had any trouble doing the work at hand.) We have Windows 7 on hand and can go back to Windows XP in a pinch, but typically when I need something I haven't committed to on my current install, I just remote to one of the workstations (or VMs) with a current stable Windows release. Admittedly, if I couldn't manage my work due to the way I test each new workstation configuration, that'd be incompetent, but it'd be equally incompetent, maybe more so, if I couldn't work without a snowflake configuration on whatever machine that happens to sit in front of me.

Where's the source?

[Anonymous Coward]

It's GPL, I want links, I want to run a supercomputer OS on my desktop

Re:Where's the source?

[Highdude702]

I'm rather sure that means since the super computer builders are building their own OS out of Linux, they don't have to supply anybody the source as their not sharing it. However I may be mistaken, anybody here that knows the contract of Linux that can verify this?

Re:Where's the source?

[Anonymous Coward]

I'm rather sure that means since the super computer builders are building their own OS out of Linux, they don't have to supply anybody the source as their not sharing it. However I may be mistaken, anybody here that knows the contract of Linux that can verify this?

Most of the programs run on super computers do not scale well down on a desktop, secondly the interfaces are often very simple input driven data crunching ones. But the complex calculations that they can do require many more cycles than desktop software and work largely based upon ideas that do not correlate well with simple desktop functions. For instance weather simulations, they crunch huge volumes of data to develop probabilities and the final outcome of all the number crunching is a simple expression of something which can be expressed with visual shapes on the screen that make computer game engines look like something really amazing. What the geek at the other end of a super computer is looking for is not smooth sword play between two waring Vikings with realistic blood spatter. What they are looking for is a picture of unforeseen consequences from changes in data.

Yes most of the algorithms and code created by scientists for super computers is open source so that it can be peer reviewed. There are exceptions especially when it comes to work that is paid for by major corporations to create new molecules and the like but by and large the core scientific number crunching code written for super computer programs is open source. Because if not how can their findings be adequately peer reviewed before publication? The average Joe with a multicore gaming monster or writing code for games is not in the same league as scientific computing and would be astonished at how shitty super computer graphics for most large data programs look.

Re:Where's the source?

[AvitarX]

If the institution using it is building it, no requirement to distribute.

If a third party is building it, they only need to provide it to the one they build I for, and only upon request.

This is GPLv2

I've requested source code from a company that had GPLv2 software and didn't give source once specifically to change the folder select dialogue for my personal use once (I'm not a real coder, but it was exciting and made my life much better).

I think they erroneously thought they had to comply with the GPL because they used ffmpeg, but they distributed it as a separate folder and ran it through the command line, so I doubt they did.

Anyway, I suspect I was one if very few people to ever request the code, it was pretty much as good as closed source (it was a utility that went along with an expensive closed source app that they would never give the source to).

Re:Where's the source?

[rgbatduke]

I'm not even sure what you are asking here. Do you truly have no idea how a GPL works?

Anyway, you have this exactly backwards. The reason Linux became popular during the parallel supercomputing "revolution" (and I say this as a modest expert, at least at that time) is because it IS an open source operating system, so you could hack the kernel, write your own kernel drivers, fix things like networking bugs or system balance issues, and handle memory at a very primitive level. You got then, and can easily get now, the complete source of the OS and all of its device drivers, although the latter has been a constant source of contention between hardware mfrs who think that a device driver that makes their hardware run is some sort of "trade secret" and the keepers of the Linux kernel. Over decades (at this point) the mfrs have largely given up and actively help with kernel drivers instead of insisting on binary-only distributions. This played a critical role in the development of early parallel supercomputers once Linux had its first kernel capable of symmetric multiprocessing with two (and rapidly more) CPUs or (later) cores, or both. That would be roughly kernel 2.0, although there were still serious issues with race conditions, (network) driver interrupts and lockups, memory management, and so on, through 2.0.4+ -- really they went on forever as the 2.0 kernel wasn't truly symmetric, handled interrupt locking "badly", and took a lot of revision and some new paradigms to smooth out and stabilize. Ah, those were the days...

Microsoft, on the other hand, made you sign away your firstborn child in order to get a copy of the OS source -- even as a research institution. If (say) your network drivers were slow, or locked up while multiprocessing, you were SOL. You COULDN'T fix it. You couldn't even find the bug. And it wasn't worth the effort -- even if you sacrificed a goat and got the source -- to learn to work with the source because it changed at MS's whim and all your work could go down the tubes at any moment and if you DID develop anything that ran on their system in some "custom" fashion, you ran into serious issues if you wanted to share it. You COULDN'T share your work with anybody else, not unless they had a surplus of goats or firstborn children too.

"Anybody" (with a need and decent programming chops) could join the linux kernel list and communicate directly with the main kernel developers and report bugs, contribute fixes or drivers, etc. There was a lot of healthy debate about what needed to be fixed, or improved, first, second, third etc, as well as just how to go about fixing them -- sometimes it required substantial redesign and had to wait for a major bump (and a lot of testing). You could of course hack/fix your own kernels or add your own device drivers, or fix broken drivers, or mess with internal "tuning", and I and many others did, but behind the public scenes the actual kernel developers -- the heart of linux, as it were -- made steady, inexorable progress.

By the year 2000, Linux had made serious inroads into not only the top 500, but there were literally uncounted small clusters that weren't fast enough (or weren't architected correctly) to crack the top 500, which relied on things like the Linpack benchmark to determine who to include. There were lots of folks who didn't USE linear algebra in their computations who built massively parallel compute farms with many different architectures and purposes who didn't even have the benchmark software installed (or give a shit) about their "ranking". Both PVM and MPI were fully ported onto Linux and most of their ongoing development was taking place on Linux boxes. Additional tools for management and job distribution and much more were developed -- on mostly Linux boxes, but yeah, there were still SGIs and Sun Microsystems clusters and much more out there. They suffered -- badly suffered, terminally badly suffered in pretty much all cases -- from being much, much more expensive than over the counter Intel or AMD box

Re:Where's the source?

[behrooz0az]

If a program depends on a gpl part in any way without any alternatives to do its primary function then it must be gpl as well. It's a clear rule.

Re:Where's the source?

[AvitarX]

Is it?

I thought if a program was only running command line functions of an executable it was not linked.

At the very least that's debatable I'd think.

Re:Where's the source?

[Anne Thwacks]

Most of the programs run on super computers do not scale well down on a desktop

Most programs run on Supercomputers are probably as old as Linux, if not older. I am pretty sure a dual processor quad-core Intel processor will beat the pants off a Cray Y, let alone a CDC7600.

It might take a while to hack the Fortran from FTN to GCC, but its a lot easier if the supercomputers are 64bit machines running Linux and not Chronos on a 60 bit machine.

Re:Where's the source?

[Tough Love]

I'm rather sure that means since the super computer builders are building their own OS out of Linux, they don't have to supply anybody the source as their not sharing it. However I may be mistaken...

You are mistaken. Top 500 shops are regular contributors to mainline Linux development, with test cases, patches and more than a few core developers. They do it because they benefit from it, and they save money that way, they don't need to carry patches. And they aren't "competing" in the commercial sense, they just want the best system they can have, and that means, play with the community.

anybody here that knows the contract of Linux that can verify this?

Contract??? You really don't get it, good luck with that.

Re:Where's the source?

[Anonymous Coward]

I can feel the excitement of a research team in early 2000s :D

Very informative! Thanks for the info!

Re:Where's the source?

[Anonymous Coward]

It depends on what the courts would decide is a "derivative work", though just running a program (especially without any pipes etc.) rather shouldn't qualify.
It usually becomes difficult when people wrap some GPL code in a DCOM interface and use it through that. Don't think any of these ever made it to court.
But FFmpeg normally is LGPL anyway, so they only need to enable you to replace the FFmpeg part of the code anyway, so even if they had e.g linked dynamically the FFmpeg code would have been sufficient.
Maybe they used other code that meant it had to be GPL, or maybe they (or some engineer in secret) just decided that making the code available was the best way to do it. Even companies developing proprietary code sometimes feel generous sometimes when it doesn't affect their bottom-line.

Re:Where's the source?

[Anonymous Coward]

The GPL is a distribution license. If they don't distribute their derivative work, they don't have to provide their source code. If they do distribute it, they only have to provide their source code to the direct recipients. Derivatives can only be redistributed under the GPL, but there is no requirement to redistribute.

Re:Where's the source?

[UnknowingFool]

He's not wrong about the GPL. You only have to publish your changes if you distribute. If you use it personally and internally, you don't have to publish your source code. However many super computing environments are academic institutions who have no problems sharing their changes and modifications. But that doesn't mean all sites must. Some might fall under competing guidelines. For example, the top US computer is Titan at Oak Ridge National Laboratory. It is a national government laboratory so there might be a mandate to publicly share information but it also does military research so some of what it works is sensitive to national security concerns. The most likely use of their supercomputer is nuclear detonation simulations. So changes to Titan to make the computer faster in general could be published (network, latency, I/O, etc). Changes that made nuclear calculations faster might not be published.

Re:Where's the source?

[AvitarX]

I suspect over caution from legal.

It's simply a DVD ripper and video converter that's main purpose is to get mpeg 1 files from any format ever for the sake of syncing deposition text to video.

They make their money selling the auto syncing at a few bucks an hour, not on a video conversion tool of the class if $30 video conversation tools.

There's very reliably took all inputsl files, handled scaling and shape correctly, and spit out a file, with zero work, so I liked it. What I didn't like was the windows default file folder picker doesn't let you pasted in a path.

If it was the engineer wanting it open, I suspect they'd make the source easier to get. It had the feeling of an "oh shit, we better add this"

Not surprising

[jwhyche]

From what I know about the windows kernel it couldn't scale upwards well enough to run in this league. And If I remember correctly one of the key goals of Linux was to make sure it could scale well on big iron systems.

We still don't know if you can successfully beowolf cluster a bunch of the old Microsoft Barnies though.

Re:Not surprising

[Anonymous Coward]

And these piece of shit Linux nodes fail all the time.

That must be why all 500 are running Linux - for the great failure rate. Or maybe you're wrong.

Re:Not surprising

[Anonymous Coward]

The kernel is, of course, important to the overall performance of the system. That's not to say that it's important that the kernel be Linux. Windows kernel will never be seen powering supercomputers because it's not open for vendors to modify, and it would take a lot more work to make it useful in super computing.

Re:Not surprising

[Anonymous Coward]

Nodes fail out constantly and have to warm restart; they typically will just restart Linux and start over on whatever dataset they were working on and so the result is a little delayed, not a big deal. But when it happens hundreds of times per second it can really screw things up. Sometimes an entire calculation can be put into jeopardy. But this is acceptable because there really isn't a good, free alternative.

Re: Not surprising

[Anonymous Coward]

What is not surprising is that a supercomputer is required to run GNOME 3 at acceptable speeds.

Re:Not surprising

[Zaelath]

So they're a $5-6 million in electricity to run, $100-200 million to design/assemble, but what really decides the OS to run is the cost.

OK.

Re: Not surprising

[Anonymous Coward]

The constant errors and huge risk must be why they all run Linux, right?

I've actually run and tuned HPC clusters with Linux - at the company who makes the CPUs in most of them (any probably the CPU in your laptop). I'm pretty sure my experience with huge stable clusters both in-house and sold to thousands of customers outweighs that of a couple Raspberry Pis without heatsinks in your mom's basement. And my experience says you're wrong.

Re:Not surprising

[Anonymous Coward]

Who says these super computers are running a free variant of Linux?

Re:Not surprising

[xvan]

Failures on supercomputers are usually hardware and not software related.
Take cosmic rays: you'll have one double bit flip on memory per day on 75000 DIMMs http://www.fiala.me/pubs/paper...
Titan has 18688 nodes.
As the amount of nodes increases, the time to first failure gets lower and lower. I heard that next gen supercomputers will have jobs maximum time reduced from current 1 day to 4 to 6 hours.

Re:Not surprising

[Kjella]

From what I know about the windows kernel it couldn't scale upwards well enough to run in this league. And If I remember correctly one of the key goals of Linux was to make sure it could scale well on big iron systems.

Originally? No, not at all. "I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones." But it's the sort of thing you can patch a kernel to do so this happened.

Re:Not surprising

[Anonymous Coward]

No the OS is literally an after thought. It’s literally the least important part of the computer.

Re:Not surprising

[Anonymous Coward]

Why do you even need an operating system. Why not run 'on bare metal'? If all that you are doing is calculating the determinant of grammians on 20kx20k matrices, what does a keyboard interface add?

https://en.wikipedia.org/wiki/Bare_machine

Re:Not surprising

[Anonymous Coward]

ever since post 2003 days windows scales just fine on the these type of systems and windows was on many of the huge core count boxes for a while, but it just doesn't make sense from a customisation perspective. these are dedicated machines to usually a very small set of tasks, you don't want a whole OS and all the gunk that goes with it and you can bet that most of these system Linux would be near unrecognisable as the key benefit is they can implement only what they need.

Re:Not surprising

[Anonymous Coward]

No the OS is basically a last minute thing and the best easiest way to create something that will work is take a system that is already running (i.e. Linux) that they have full source code access to and then cut out the bits they need. what ends up running has very little resemblance to Linux in anything but the name. remember 5 or 10 million is going to buy you fuck all when it comes to developing an OS so unless they are willing to commit an extra 100 million or more then they have to cobble together something from what is available.

Re:Not surprising

[Anonymous Coward]

for the many of them saying they run Linux is really BS, all they have done is taken a few bits that they needed and saved themselves some coding time, they don't run Linux in anything but name.

Re:Not surprising

[Anonymous Coward]

It's the cost of support staff. Some of the HPC providers would throw in a free office block with on-site engineers with every supercomputer purchased with a support contract.

Re:Not surprising

[AmiMoJo]

The main issue with Windows is not scaling, it's been able to make good use of >100 cores since the early 2000s. The issue is management.

These computers don't run one single OS. They run multiple copies of the same OS on nodes, and dispatch work to those nodes using special high speed interconnects. When you have thousands of CPUs, power supplies, RAM modules etc. some of them are going to fail, so you divide them up into nodes that can fail and recover individually.

Windows is not well suited to this. With Linux you can just create an OS image and deploy it over a network. Windows can kinda do that, but it's awkward and not really suited to embedded, headless operation in a highly networked environment.

Re:Not surprising

[Anonymous Coward]

Aaahahaha you've got no idea what you're talking about. Totally have no idea how resilience is handled.

> hundreds of times per second

Wow, you're totally out to lunch. Absolutely no clue about the technologies or scale involved.

No, supercomputers typically perform periodic application-level checkpointing, and a failure will result in the entire application being brought back up from the last checkpoint. They don't just bring nodes back in to rejoin ongoing computation (that's an area of research and development, but not how most of today's codes work).

And mean time between failures across the cluster might be days or hours or minutes (if it's a particularly unstable system). It's not milliseconds. You're only at least 5 orders of magnitude off there.

Re:Not surprising

[Anonymous Coward]

So they're a $5-6 million in electricity to run, $100-200 million to design/assemble, but what really decides the OS to run is the cost.

OK.

Don't under estimate the hassle of dealing with licensing. :)

Being "in compliance" can take a full time job and the risk in fines is worth considering.

Re:Not surprising

[JanneM]

Even if you could create a version of Windows that would run well in a cluster environment, the applications aren't there. One reason Linux is so dominant is that the kind of code you want to run on a big cluster is all written for Linux.

With that said, some HPC apps are hybrid; you run a client on your local machine (Windows app, or a web app) that dispatches the work on to the cluster in the background. With those you can treat the cluster as a big accelerator for your local app.

Re:Not surprising

[Anonymous Coward]

No the OS is literally an after thought.

No.

Itâ(TM)s literally the least important part of the computer.

No.

You are either trolling, really stupid, or both.

Re:Not surprising

[Tough Love]

Who says these super computers are running a free variant of Linux?

I do. There is no such thing as a non-free variant of Linux.

Re:Not surprising

[Tough Love]

Failures on supercomputers are usually hardware and not software related

Specious. Failures on supercomputer bring-up (where the big frontend engineering costs live) are usually software and not hardware.

Re:Not surprising

[Tough Love]

Why do you even need an operating system. Why not run 'on bare metal'?

Because each node needs to manage its hardware resources, interact with other nodes, and provide a standard binary api to applications. In other words, it needs an operating system.

Re:Not surprising

[Tough Love]

for the many of them saying they run Linux is really BS, all they have done is taken a few bits that they needed and saved themselves some coding time, they don't run Linux in anything but name.

You are an idiot.

Re:Not surprising

[Anonymous Coward]

Or I actually work on these system at NERSC and OLCF and know what I’m talking about.

Are the supercomputers fully backdoored?

[Anonymous Coward]

Linux security has been proven lately to be quite weak... Who knows how many of these supercomputers are in somebody's botnet...

Re:Are the supercomputers fully backdoored?

[stooo]

Yes.
Supercomputers usually have front and back doors. It's needed for maintenance.
So yes, they are alwas backdoored.
https://media2.s-nbcnews.com/j...

Re: So I guess this proves *BSD is dead

[Anonymous Coward]

triumph of mediocrity!

What a pathetic bunch of comments so far

[GerryGilmore]

How far has the discussion quality fallen? Apparently this low, even without a political bent.

Re:What a pathetic bunch of comments so far

[DCFusor]

Sorry, out of mod points...not that they help that much when we're swarmed by immature idiots and AC trolls.

Re:What a pathetic bunch of comments so far

[Freshly Exhumed]

I signed up here almost from the very beginning of Slashdot time. A few days earlier and I might have had ID#100000. From what people have been saying here in all those 19+ years, Slashdot was dying, has always been dying, is still dying, and will always be dying. I just ignore the irritating cruft, and I'd advise everyone to do that.

So, moving right along... how about those TOP500 scores!

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

How far has the trump quality fallen? Apparently this trump, even without a political trump.

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

Four percent of posts on Slashdot are about how much better Slashdot used to be at some unspecified period long ago. Another 9 percent are about how much better the IT industry used to be.

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

The submission is not exactly very controversial or technical enough to raise much discussion at the deeper end of the pool. Policy discussions are of course left out so let me begin by stating that according to the Internet, US exascale strategy focuses on usable performance with actual codes, not just to win at the top500 list. So there, the US people worrying about US rankings in the world can rest easy. Your government hasn't yet lost it's marbles in this respect.

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

How far has the discussion quality fallen? Apparently this low, even without a political bent.

i fukd ur mom

Re:What a pathetic bunch of comments so far

[Steve72]

Imagine a Beowulf cluster of these?

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

Slashdot is still dying, Netcraft confirms it!

Re:What a pathetic bunch of comments so far

[Anonymous Coward]

Eh, slashdot or any other site is mostly trolls. I actually believe, or want to believe, that most of the trolls are actually intelligent folks that could contribute but see the discussion as trite and post ridiculous crap for their amusement. I was young once too.

Re:What a pathetic bunch of comments so far

[CBravo]

In Russia, Beowulf cluster imagines you.

Re:What a pathetic bunch of comments so far

[AbRASiON]

Can you elaborate on what you think the problem is?

I haven't noticed them being particularly bad, however I have noticed there's less and less. Particularly the last 4 months or so? Has the gradual decline sped up?

Re:What a pathetic bunch of comments so far

[AmiMoJo]

Slashdot has definitely changed. I remember when it was more of a marketplace of ideas, where interesting comments were actually modded "interesting" instead of "flamebait" or "troll". I remember when everything wasn't a conspiracy of some kind.

Back on topic, imagine a Beowulf cluster of the top 500 supercomputers!

Re:What a pathetic bunch of comments so far

[JanneM]

Todays supercomputers and clusters are all descendants of the Beowulf clusters. No need to imagine anything :)

Re: What a pathetic bunch of comments so far

[Duncan J Murray]

Popped back here after a few years away. My thoughts exactly.

Re:What a pathetic bunch of comments so far

[Steve72]

Fair point.

How about imagining Natalie Portman naked and petrified? ;)

Microsoft's supercomputing efforts

[Freshly Exhumed]

As Linux began to crack the TOP500 list in the 1990s, Bill Gates tried to ignite a supercomputer effort at Microsoft but it never amounted to much. I wish I could find a link to it. Anyways, I found the following timeline for Microsoft's "Project Catapult" AI-related supercomputing effort, which might not be in the TOP500 list's league:
        2010: Microsoft researchers meet with Bing executives to propose using FPGAs to accelerate Indexserve.
        2011: A team of Microsoft software engineers and researchers come together to address a huge processing problem: how to use customized, programmable integrated circuits to accelerate computationally expensive operations in Bing’s Indexserve engine.
        2012: Large scale pilot of FPGA boards in each of 1,632 servers and wiring them with a custom secondary network.
        2013: Results of pilot demonstrated positive ROI, allowed latency improvements in ranking while cutting the number of required servers in half. Decision was made to go to production.
        2014: Publication of paper and decision to merge Bing design with Microsoft’s converged SKU, adding to the v2 architecture that enables configurable clouds.
        2015: Ramp up to large-scale production in Bing and Azure.
        2016: “Configurable Cloud” architecture in nearly every new production server. Configurable Cloud paper published (Micro 2016, October)
https://www.microsoft.com/en-u...

Re:Microsoft's supercomputing efforts

[Bert64]

There was an "HPC" edition of windows 2003, and microsoft managed to sponsor a few places to build clusters using it that made it into the top500 list...
I don't recall anyone ever using it of their own volition tho, only if microsoft were paying, and at least one of those clusters was a dual boot experiment which climbed 50 places in the ranking when booted to linux.

Re:Microsoft's supercomputing efforts

[Anonymous Coward]

Yes and only for the press release. The hpc department at my university built a new cluster during the time when i studided there. They got alot of sponsor money from microsoft and installed windows 2003 on all nodes. Microsoft got their press release and then the univeristy wiped all nodes and installed linux.

That cluster was on the top 500 list at the time.

Re:Microsoft's supercomputing efforts

[Tough Love]

one of those clusters was a dual boot experiment which climbed 50 places in the ranking when booted to linux

Really, this is the only comment anybody needs to read in the "why not Windows?" thread.

Obligatory xkcd

[Gravis Zero]

Some features are used more than others. ;)

Re:Obligatory xkcd

[Freshly Exhumed]

That time when Linus's wife could not be Rickrolled because her Linux box had no Flash capability was a searing tragedy in the annals of computer history.

Makes sense

[Anonymous Coward]

It's easy to tune something for a specific use when it doens't do much else.

So, Linux turned the Top500 into a Monoculture

[williamyf]

So, the Top 500 list of computers was dominated by many Variants of Unix, with a little sprinkle of other weird stuff (among those, VMS). Which is not a monoculture

Then, as the other weird stuff waned, Windows took it's place (for a short while). Not directly as a replacement of course, but rather as a percentage of Top500 systems.

On the other side of the fence, Linux began to take increasign market share of the Top500 because of low cost, shallow learning curve from *nix, and posibility to modify source code, in an accelerated path to become a monoculture (at least where the Top500 is concerned).

And now, finally, we are on a monoculture in the Top500, with Linux all the way in the Top500... No *BSD, no AIX, HP-UX, or Solaris. Just Linux all the way.

Better not catch anyone complaining about Chrome Monoculture, Windows Monoculture, or Android monoculture! M'kay? ;-)

Re:So, Linux turned the Top500 into a Monoculture

[Freshly Exhumed]

These latest TOP500 project teams seem to have exercised free choice of OSes, free from force or coercion. Choice is good, so I'm not sure if you're complaining, and if so, why? If we knew that all 500 projects were using completely interchangeable code and hardware we might have a monoculture at play, but the reality is that they used the best available OS option for their own specific, bespoke, custom needs. I hope I understood your comment correctly.

Re:So, Linux turned the Top500 into a Monoculture

[quantaman]

So, the Top 500 list of computers was dominated by many Variants of Unix, with a little sprinkle of other weird stuff (among those, VMS). Which is not a monoculture

Then, as the other weird stuff waned, Windows took it's place (for a short while). Not directly as a replacement of course, but rather as a percentage of Top500 systems.

On the other side of the fence, Linux began to take increasign market share of the Top500 because of low cost, shallow learning curve from *nix, and posibility to modify source code, in an accelerated path to become a monoculture (at least where the Top500 is concerned).

And now, finally, we are on a monoculture in the Top500, with Linux all the way in the Top500... No *BSD, no AIX, HP-UX, or Solaris. Just Linux all the way.

Better not catch anyone complaining about Chrome Monoculture, Windows Monoculture, or Android monoculture! M'kay? ;-)

I think the reason it's become a Linux "monoculture" is that it isn't really a monoculture.

Top-500 should be an area that's amendable to variety. Any one project is big enough that some serious customization is going to occur, so traditionallyany one OS could focus on a specific feature set and nab themselves a bit of the market. That's why the big Unixes co-existed for so long, if your problem was a round hole you'd grab the Unix that looked the most like a round peg, and if you had a square hold you'd grab one that looked like a square peg.

But Linux is modifiable, so if you need a round peg you can make it a round peg. If you really need a specific feature or optimization you can write a kernel patch. That doesn't matter for the consumer market but it matters for the top-500 list.

I wonder what the future of the BSDs is, they have cool communities but I'm not really sure where their niche is.

Re:So, Linux turned the Top500 into a Monoculture

[Anonymous Coward]

I suppose it's all about cost for the proprietary OSes, but what of the BSDs? They're free, and the license would let custom work on a super computer's OS be closed and even sold. Is it networking speeds or the like? Parallelism? I would really like to know, didn't come here for "'cause Linux Rulz!" ass-hattery

Re:So, Linux turned the Top500 into a Monoculture

[quantaman]

I suppose it's all about cost for the proprietary OSes

Not for the top-500, I'm sure Linux is cheaper that the Unixes, but you don't get a machine on the top-500 by cost-cutting.

but what of the BSDs? They're free, and the license would let custom work on a super computer's OS be closed and even sold. Is it networking speeds or the like? Parallelism? I would really like to know, didn't come here for "'cause Linux Rulz!" ass-hattery

I think it's simpler than that, what does BSD have that Linux doesn't?

Linux gives you an open source Unix with a massive community and a ton of corporate backing.

BSD gives you an open source Unix with a small community and a little bit of corporate backing.

I'm sure there's some specific application where BSD might have an advantage, but there's going to be a lot of applications where Linux has big advantage just because there's so many technical resources thrown behind it and so many skilled developers and admins available.

It's not so much a criticism of BSD, it could have gone differently (though it probably would have needed a copyleft license to attract the community) but the open source community has largely unified around Linux.

Re:So, Linux turned the Top500 into a Monoculture

[Anne Thwacks]

Top500?

We are talking about Supercomputers here. Surely you mean TOPS20!

Re:So, Linux turned the Top500 into a Monoculture

[coofercat]

So you're saying people are using Linux as a 'framework' upon which they build their own (custom) supercomputer OS? Nice! :-)

Re:So, Linux turned the Top500 into a Monoculture

[williamyf]

[...] so I'm not sure if you're complaining, and if so, why? [...] I hope I understood your comment correctly.

Right at the end of the comment, there is a ;-) emoticon.

You may have missed it.

Re:So, Linux turned the Top500 into a Monoculture

[afidel]

The BSDs I run into mostly in embedded applications, mostly security and network appliances. The primary drawback to the BSDs in my experience is driver availability and that doesn't really apply if you're building your own appliance so long as your upstream providers can be convinced to supply a BSD driver for your application.

On the other hand...

[hyades1]

"All 500 of the World's Top 500 Supercomputers Are Running Linux"

And they still can't get their graphics cards to work right.

Re:On the other hand...

[nyet]

Who is "they"?

Re:On the other hand...

[hyades1]

Who do you think.

Re:On the other hand...

[Anonymous Coward]

A supercomputer with a graphics card? WTF are you talking about?

Re:On the other hand...

[hcs_$reboot]

Haven't had, neither other Linuxers I know, any graphics card problem for a long time.

Re:On the other hand...

[Bert64]

Several of the top500 are using GPUs, but for calculations rather than displaying graphics. Having an active video display on a large cluster would be stupid, most supercomputer nodes won't have screens attached and while the power consumption of an idle display controller is pretty low its not 0, and multiplied by thousands of nodes its a terrible waste of power.

Re:On the other hand...

[Jeremi]

Do supercomputers ever come with a graphics card (that is intended to drive a display)?

I'd imagine that if you want a console for your supercomputer, you set up a PC next to it and run X Window remotely or something.

Re:On the other hand...

[Anonymous Coward]

Do supercomputers ever come with a graphics card (that is intended to drive a display)?

I'd imagine that if you want a console for your supercomputer, you set up a PC next to it and run X Window remotely or something.

In most cases a simple terminal will do nicely. Xwindows is only necessary if you are running imaging, there are stripped down versions of X that can do everything graphic necessary. If you are talking about something marvelous like sliding through a human body in details down to fractions of a millimeter from an advanced medical imaging device and simulating disease progression then the graphics has to be turned up a notch. Simulating a complex set of changes to any graphical data like the structure of a human body requires much more number crunching than just what is seen on the screen. So the graphics is the final step in the number crunching done in complex simulation and by and large does not need to happen in real time the way say game play and high definition video does. The final product of a simulation can be a high definition video but the primary work of supercomputers is simulation done through number crunching.

Supercomputers are great at creating mountains of data but the data still needs to be interpreted in some way and real time graphics pushed to a human brain to try to interpret mountains of data can very easily become an insurmountable bottle neck unless you are Mr Spock ;-). Most scientists use time on supercomputers only for complex statistical analysis of data not video data though often the end product of their work is represented as a video simulation to help explain what they are working on. man man, man grep, man cat, then get back to me.

Re:On the other hand...

[Miles_O'Toole]

I believe some do, though they don't necessarily use it to put images on a monitor.

Re:On the other hand...

[dargaud]

Some do, to run CUDA on it. But often it's not standard graphic cards, but derived hardware such as Xeon Phi.

Re:On the other hand...

[Anonymous Coward]

Node on a supercomputers aren't used exclusively by one application all the time. Each has gigagytes of memory, a multi-core CPU and GPU cores. A project will be scheduled to use a particular number of nodes for a fixed amount of time. So, there could be multiple users doing different things. You have to gather all that data from those different nodes and send it to a single display system to render.

Re:On the other hand...

[niks42]

TBH, where GUIs are desired for cluster management &c, it's often easier to present an HTTPS service on the nodes (or head nodes) so you can point a browser at it. I am working at the moment with an HPC build in a hospital; I have to use their VPN solution to get into the cluster, and that routes me to a Windows box to act as a jump-off point with an account on the local AD. Of course the Windows box has a web browser (ancient) - so that comes in quite handy. The hospital IT department won't install an X server on it, so thankful am I that most of the GUI tools can be run from a browser.

I had to find a terminal emulator that I could use without installing, so Teraterm came to the rescue again. Lovely program that stores its settings in the directory it is launched from as well .. There are still some cluster tools that either require X or a client that needs installing (more sucking of teeth from the IT department. I found one set of cluster tools launches Firefox and provides a plug-in for it, but IT won't install Firefox.I can't find an X server that doesn't need to be installed on the Windows box either.

But...

[Anonymous Coward]

Can it play Crysis?

Re: Irony

[Anonymous Coward]

well, your last name is âoebigdickâ so we know what you are (not) packing

Yeah. And?

[thedarb]

What'd you expect it to run? Windows?

Distro

[ShakaUVM]

Anyone have any information on what distro they use? The article didn't say.

Re:Distro

[Anonymous Coward]

Cray supercomputers run CrayLinux.

Others... Red Hat has a presence.

Re:Distro

[Freshly Exhumed]

I'd expect that most of them are not distro-based but rather LFS-based: http://www.linuxfromscratch.or...

Re: That's because... You are Steve Balmer

[Anonymous Coward]

That or Linus ran over your dog or some other heinous offence.

Re:Yeah. And?

[Freshly Exhumed]

I'm sure some of those rigs could spare a few CPU cycles to run VMs in case somebody needs to Skype their basement-dwelling maladroit kid.

Re:Yeah. And?

[iggymanz]

five years ago, 3 of the top 500 did run windows, and in 2011 4 did.

Verily, it is the year of the linux supercomputer

[Anonymous Coward]

The year of the linux desktop can't be far behind!

Linux with systemd --- bleak future ahead

[Anonymous Coward]

The development of Linux seems to take a sharp left turn some time ago, with the advent of systemd

My question to the real /. folks will be, is this the future for Linux you like to see?

Re:Linux with systemd --- bleak future ahead

[Anonymous Coward]

The development of Linux seems to take a sharp left turn some time ago, with the advent of systemd

My question to the real /. folks will be, is this the future for Linux you like to see?

Fortunately to create a supercomputer systemd is completely unnecessary. The people who build them are capable of developing their own inetd and have for many years done exactly that. Systemd is for those who cannot create a daemon and do not have a clue as to what the boot sequence of linux is or how it is read by the kernel at boot time. This is where Linux shines you can create your own boot environment and have access to all the documentation and code you need to do the job.

Supercomputers require special configuration knowledge and this has been the case all the way back when most of them ran some form of Unix, mostly Sun, IBM or HP. With the death of AIX, HP Unix, and Oracle essentially murdering Sun, Linux has taken over the field because it does not require that you use any specific init system especially systemd which in structure tries to clone how OSX boots. I will say this: it does obscure the boot sequence from the user unless they read the boot script and code of the particular "system daemon" in their distro and learn how to modify it thus creating their own OS based on Linux.

In some ways coding for systemd is easier than learning which rcd to modify but then again the skill of triggering a kernel read and knowing which run level your process should run at is an essential skill in developing a super computer running any operating system based upon the Linux kernel. So learning how to code for systemd is not all that evil it is just another layer of complexity added upon inet. Coding for supercomputers systemd can be used but is certainly not essential and the people who are best at building them can and do build their own os runtime base on the Linux kernel without help from Lennart or Red Fat.

Re:Linux with systemd --- bleak future ahead

[jabuzz]

Though I think you will find that the vast majority are running some mainstream Linux distribution on the nodes. Whether that is a RHEL derivative (CentOS/Scientific Linux) or a LTS version of Ubuntu etc. if it's latest its systemd.

I guess Linux won.

[Anonymous Coward]

Today I saw a strange street multimedia panel on the street (actually seemed to be a flat panel TV in portrait orientation).

You those which always are shown in photos with a Blue Screen of Death?

Well, that one had a lot of chars and when I came close I could read "Warning", followed by the usual messages we see in boots when Plymouth does not hide them.

Linux replaced crashed Windows panels in ridiculous public displays of shame. That was the last bastion for Windows -- and it fell...

Any cluster scales

[Anonymous Coward]

I run a cluster of Android tv boxes FFS, running my particle simulations. You can get anything to scale if you can network it and there are plenty of Windows clusters doing stuff.

However, who the f*** wants a GUI OS, running data crunching? They don't have a clean core there to scale separate from the GUI.
Can you imagine 10000 cores running popup windows reminding you to download the latest virus definition???

No thanks.

Re:Any cluster scales

[Anne Thwacks]

plenty of Windows clusters doing stuff.

The technical term for this is "botnets".

Re:Any cluster scales

[Anonymous Coward]

Windows IoT

Re:Any cluster scales

[niks42]

>> Sent from my ASR33 using ASCII

I didn't have lower case on my ASR33 .. you use embedded hex codes to type that?

Re:Yeah. And?

[hcs_$reboot]

What'd you expect it to run? Windows?

Some people would expect that. But the Linux kernel is certainly more customizable than a Windows black box (that would require the help of Microsoft engineers).

Telemetry?

[Anonymous Coward]

Maybe all the owners of these supercomputers don't want Microsoft snooping around on them.

Nobody will care one day soon.

[Anonymous Coward]

As if in 10 years time anyone is going to know or care what OS their functions are running on - except old geeks in rocking chairs.

Re:Nobody will care one day soon.

[thinkwaitfast]

As read ~27 years ago amidst the OS flame wars: The only people who should know about operating systems are programmers. Users do not interact with operating systems

Re:Nobody will care one day soon.

[Anonymous Coward]

If you replace "operating systems" with "kernels" that's still pretty accurate, and on Linux it's even true if all your hardware is supported by in-tree drivers. Install, boot, and use your hardware.

Re:Yeah. And?

[rgbatduke]

And, I'm betting that Microsoft sponsored all of them, just to have SOMETHING ON THE LIST. But did M$ ever manage to bribe enough people to get 1 lousy percent of the top 500?

For most people, the extra HUMAN expense of making a cluster work at all, and the extra TIME expense of having it run like a pig when you get it to run at all, isn't worth even a massive M$ bribe free cluster (as long as you run Windows). It sort of depends on whether actually getting your work done is more important to you than pain.

BSD

[ArchieBunker]

The *BSDs will keep on doing what they have always done. Run well with minimal upkeep and not beta test features on production releases. Under Linux the mentality is if something compiles then ship it. I ran Linux in the 2.0.x kernel days. What they call Linux today is so far removed it might as well be a different operating system. Some distros don't even include tools like nslookup or traceroute anymore. Good luck installing that package if your default route isn't set. Oh and "route" has been changed to something else now for no good reason. What exactly was wrong with the old program and syntax?

Re:BSD

[Anonymous Coward]

It wasn't maintained for years is what was wrong with it.

Re:BSD

[swb]

I switched to FreeBSD from Linux ages ago because it was a complete system. Linux was a kernel with a bunch of utilities bolted onto it, and there was That One Day where I was trying to upgrade something and needed a key utility for configuring something and it wouldn't run, and there was no "source" for an updated version. I gave FreeBSD a spin and just liked that everything was a part of a larger whole, and not a bunch of pieces with varying standards.

FreeBSD can have other problems, sometimes certain ports are broken for a long time, for example, but generally speaking the base system always works.

Ironically, I think that "the Linux desktop" actually has contributed to Linux large OSS market share, despite it never quite being the year of the Linux desktop. I think that when distros emerged with out-of-the-box GUIs and browsers a lot of technically oriented people who might not have run a CLI-only OS were able to "get into" Linux because the GUI was just good enough to be basically usable, even if it wasn't feature complete with Windows.

Re:BSD

[ArchieBunker]

What needed to be maintained exactly? Were there bugs or security problems? If the answer is no then why fix something that was never broken?

Re:BSD

[Anonymous Coward]

It's still there, it can still be used.
The old program and syntax didn't work for newer and advanced features, and it duplicated A LOT of code, so it is now all maintained in the "ip" tool. Which is not even remotely new either.
There were 2 programs doing the same thing, and the one that the distributions considered crappier was dropped from default installs. What's the big deal with that? Happens all the time.

Of course it requires a supercomputer

[93 Escort Wagon]

... to fully appreciate all the features of the latest Enlightenment desktop.

Re:Of course it requires a supercomputer

[Anonymous Coward]

There are still PCs that don't run enlightenment? My shitty pc from 2009 without videocard runs englightenment with all whistles and bells active...

Re:Of course it requires a supercomputer

[Anonymous Coward]

it's funny if you ever tried running Enlightenment in the 90's.
these days enlightenment is considered light-weight.

Re:Of course it requires a supercomputer

[93 Escort Wagon]

Hey, old jokes are the best jokes!

Re: Irony

[Anonymous Coward]

What does Ãoebigdickà mean in ASCII?

Limitation of a single computer

[Prien715]

For most parallel problems, it's possible to divide them and send each piece to different computers, rather than a different core on the same computer. For even more highly parallel problems, using a GPUs to do the computation is even faster.

With 100 gig ethernet, we're starting to see networking speeds closer to bus speeds on motherboards themselves and it's cheaper, faster to scale (especially dynamically), and probably more fault tolerant (node fail? Send the job to a different node) to use more computer nodes rather than using more processors in a single computer.

Distributed computing has almost made supercomputers irrelevant -- except for people with a hole in their pocket. Folding@home is more powerful than anything on their list while we have no idea what monster of a compute clusters work inside Google or Facebook -- but given the open source software they have released (e.g. Facebook's 360 degree video stitcher) and how slow they are on a single machine -- the only way they'd be usable on their site is if you have a massive cluster.

Re:Limitation of a single computer

[serviscope_minor]

Distributed computing has almost made supercomputers irrelevant

Not really, no. supercomputers are distributed computers with good interconnects. For many calculations, the interconnect is really, really REALLY imortant which is why a good number have the interconnect right there on die with the CPU.

Re:Limitation of a single computer

[Khashishi]

You are dreaming if you think every scientific simulation can be gamified and distributed to the masses.

FreeBSD

[sremick]

I miss the days when the list had a ton of FreeBSD systems. To this day, it remains my preferred OS. Two little software compatibility issues prevent me from running it as my desktop OS anymore although I did for many years. It still has a home on several servers here in my house where it has distinct advantages over Linux.

Re:FreeBSD

[Anonymous Coward]

Does BSD run on your "boxen" in your mom's basement, errrr, I mean your bedroom?

Re:FreeBSD

[sremick]

I'm a home-owner in my 40s, with multiple servers and network switches in our home, but nice troll-fail.

Re: So I guess this proves *BSD is dead

[stooo]

Yeah, but Triumph of the Free.

Devuan & antiX

[Anonymous Coward]

http://without-systemd.org/wiki/index.php/Main_Page

Simulated Universe

[Anonymous Coward]

So we should hope that whoever is running the simulation that we inhabit chose Linux for their OS

Windows subsystem for Linux.

[140Mandak262Jamuna]

Amazon, Google and Microsoft have cloud platforms. Technically by the "network is the computer" mantra from Sun Solaris days, these are supercomputers.

There Linux compatibility is so essential to get a toehold there Microsoft had to support linux way of doing things. Finally it relented and introduced "Linux subsystem of Windows" support.

Does it support incoming ssh connections? I use ssh to go out of Windows to connect to Linux machines in my network. If the linux subsystem allows incoming ssh and RSA keys, I see Active Directory losing, eventually, to Linux based authentication servers.

On second thought, DON'T ask him

[Anonymous Coward]

If you ask Linus, "Of course it's %^&*($#@! fun!"

Naaaa, that wouldn't be his answer

Xyrus, agreed 110% (been saying that...)

[Anonymous Coward]

Agreed, 110% Xyrus & I've been pointing it out for years now e.g. https://news.slashdot.org/comments.pl?sid=11351163&cid=55541351/ & talk's cheap (except them "$" does the talking especially for keeping per unit costs down avoiding licensing fees - especially w/ something that's free & WORKS well... which yes, Linux does ESPECIALLY ON SERVERS (& decent on the desktop too, w/ some 'caveats' listed below)).

* The ONLY THING "holding Linux down/back" on the desktop was APPS FOR PARTICULAR PURPOSES (that Windows fills the niche on already - IF those applications were either ported OR if a competing one did the job as well + could handle the leading competitor's data formats? It'd be a different story - @ least SOMEWHAT on the desktop).

APK

P.S.=> Put it THIS way - IF I didn't have a valid license for Windows 7 64-bit? I'd be on Linux, for sure (I like it well enough & have spent @ least a year on it in the past using it) CO$T of licensing Windows WOULD HAVE BEEN "TALKING" TO ME TOO as an end-user consumer (@ this point @ least - Linux used to really 'suck' vs. Windows - not anymore, UNLESS it's for a particular app for a particular purpose that ONLY Windows has & Linux doesn't - in fact, that is what I have seen drive those who TRY Linux away in the end vs. Windows (what most folks grew up on really & got used to))... apk

Re:Xyrus, agreed 110% (been saying that...)

[barbariccow]

Linux also has a shorter path to the hosts file than windows. That's gotta be a plus for you, right?

Unix desktop (i.e., Mac OS X)

[Anonymous Coward]

Unix never made inroads on the desktop.

Apple's Mac OS X would beg to differ:

* https://www.opengroup.org/openbrand/register/apple.htm

Certified UNIX(tm) since 10.5 ("Leopard") in 2007:

* https://www.opengroup.org/openbrand/register/brand3555.htm

Re:Unix desktop (i.e., Mac OS X)

[Anonymous Coward]

True, but they had to sneak it in through the side door by making it look like a Mac. :)

I think GP is saying Unix failed at door-to-door sales.

James Dean says (lol)... apk

[Anonymous Coward]

I haven't ported it to Linux & was actually hoping to OpenSORES it (until morons here threatened to create an EFast Google Chrome malicious doppleganger out of it IF I did - their loss) via Delphi X (does ALL the major platforms, yes Linux + MacOS X too) https://www.embarcadero.com/products/delphi/ so others could & there ya are...

* Left you another message (see subject & https://news.slashdot.org/comments.pl?sid=11351783&cid=55554475/

APK

P.S.=> R o T f L m A o (but the Eagles put out what I meant BETTER than I EVER could - sayin' it all "SO CLEAN" so you "KNOW JUST WHAT I MEAN" (lyrics from the tune used))... apk

Re:James Dean says (lol)... apk

[barbariccow]

Why not? Why is it so hard to port, at least a cli version with no options.. the hosts file is the same format, hopefully you're using the standard open/malloc/close file handling, which is already portable. Or if you for some reason wrote it in .NET for whatever reason, there's tools like mono and silverlight that should make easy it porting. Hell, here's a simple implementation in bash:

#!/bin/bash
BLOCKED_HOSTS="blah.example.com ads.google.com etc.whatever.com"

die() {
echo "$@" 2>&1
exit 1
}
[ ! -w "/etc/hosts" ] && die "Error: Cannot write to /etc/hosts"
for blockHost in ${BLOCKED_HOSTS};
do
if ! ( grep -q "^.*[\t ]*${blockHost}" "/etc/hosts" >/dev/null 2>&1 ) then
echo "Blackholing host ${blockHost}";
printf "%s\t%s\n" "127.0.0.1" "${blockHost}" >> /etc/hosts
fi
done

Just replace BLOCKED_HOSTS with a space-separated list of hosts you want to block, it will scan if they are already blocked and block the ones that aren't in your list. I hereby donate this code to the public domain, you're free to use it however you please.

Here's example of it working:

[root@MYHOSTNAME ___brIj4s05]# chmod +x doit.sh
[root@MYHOSTNAME ___brIj4s05]# ./doit.sh
Blackholing host blah.example.com
Blackholing host ads.google.com
Blackholing host etc.whatever.com
[root@MYHOSTNAME ___brIj4s05]# ./doit.sh # Note that the second run doesn't block anything that isn't already blocked
[root@MYHOSTNAME ___brIj4s05]#

Re:James Dean says (lol)... apk

[barbariccow]

Acftually, make the grep line this for more betterer:
if ! ( grep -q "^127.0.0.1.*${blockHost}" "/etc/hosts" >/dev/null 2>&1 ) then

Re: James Dean says (lol)... apk

[Anonymous Coward]

You just put APK out of business. How much do we owe you?

Re: James Dean says (lol)... apk

[barbariccow]

Nothing. I donated it to public domain. As an open source developer, my thanks comes from replacing black box crap with well-written, supportable, and customizable software. :)

Succinctly

[go-nix.ca]

"World Domination. Fast" -- Linus Torvalds

ssh and scripting

[Khashishi]

Having used supercomputers for some physics simulations, I have some understanding of the general workflow. There is no need for GUI nonsense on a supercomputer. You have thousands of users using the supercomputer remotely. You ssh in to submit jobs into a queue. You process input files and do some file manipulation. You write scripts to copy and edit input and output files to organize your test cases. You copy what you need to your local workstation so you can visualize the results. This is exactly what POSIX is for.

I suppose POSIX for Windows exists, but it is far behind Linux. What's the point in running Windows if you are going to use POSIX?

Generally, you want to compile your codes for each supercomputer you run on, so a full make environment and portable API is kind of important.

I never said it was 'hard to do'... apk

[Anonymous Coward]

See subject: You're trying to put words in my mouth I never said - all it would be is driveletters vs. mounted devices (ez) & some WinSock2 vs. *NIX sockets (all those abstractions ARE done already though) & of course, anytime I used Win32/64 API calls (would need analogs in *NIX variants is all - & they ARE there).

Oh yea - .ini file read/write/flush/close I use would have to be changed to std. file open/read-write/flush/close.

(That's about it though)

Charmode/tty term/DOS windows stuff? Hell no - I started it that way (3 diff. apps & an access database engine) - that's NOT what folks want today (GUI world out there now MOSTLY)... admins like it. Normal users tend not to.

APK

P.S.=> ... & "there ya go" but as I've said here @ least a DOZEN TIMES NOW? I am NOT in the "habit of helping 'the competition'" (big Windows fan here - changed my life for the better & gave me a career (but by the same token, I actually LIKE & have used Linux (1st time 1994 slackware (sucked), Redhat in 1999 (got better but still sucked) & 2010 KUbuntu (VERY GOOD))... apk

LMAO - others here tried 5x & FAILED... apk

[Anonymous Coward]

See subject: Here is a classical example (where I had to correct errors they made using shellscripts 5x https://yro.slashdot.org/comments.pl?sid=7917899&cid=50450951/ where I had to correct what they missed in data errors using 14++ *NIX commands THEY DIDN'T WRITE THEMSELVES (whereas by way of comparison I literally did into 1 consolidated stand-alone TRUE non-interpreted multitasking/multithreaded executable that users here LIKE & USE, even Malwarebytes hosting + RECOMMENDING my work (all mine no less unlike SCRIPT KIDDIES above, lol!)

Heck - Even my nephew on suggestion from me during his time @ RIT also using Python (did a decent job but lacked errtrapping etc. which I showed him & he thanked me for to THIS DAY on the job professionally) to up his skills in coding (He works in other areas @ Apple now but does scripting a lot to do so).

APK

P.S.=> "Putting ME 'out of business'" won't be doable by mere "script kiddies" - a TRUE coder in say, C/C++ might though (or Delphi Object Pascal as I did it in) so it goes to ALL platforms non-interpreted SOLID high-quality & HIGH performance in GUI (What people REALLY mostly want)... apk

Re:LMAO - others here tried 5x & FAILED... apk

[barbariccow]

what? That's a simple port I gave you in like 8 lines. It's not "Script Kiddie" as bash isn't a scripting language... it's a batch language. And as I showed in the bottom it works. Use it if you want, don't if you don't want. If you wanted to port your thing to Linux you just have to put the database into BLOCKED_HOSTS, or even better if you have a url to host the space-separated list, change that line to:

BLOCKED_HOSTS="$(curl https://www.apk.com/BAD_HOST_L... 2>/dev/null)"
[ $? -ne 0 ] && die "Failed to fetch new host list."


You don't need sort or uniq or any of that junk... I again can't tell if this is trolling or not, but as you see it already handles not adding duplicate entries via the grep conditional.
Or maybe this is a fake apk troll I dunno.. whatever.

Re:LMAO - others here tried 5x & FAILED... apk

[Anonymous Coward]

bash isn't a scripting language... it's a batch language

You write shell scripts in it, but it's not a scripting language... I see.

Please tell me more about this nitpicking thing.

You miss what other 'scriptkiddies' always do

[Anonymous Coward]

1.) wget or curl to obtain the data from remote sources (big blunder)

2.) Did you account for capital letters (many sources fuckup that way creating duplicates)? No. tr needed.

3.) Did you account for trailing blanks OR other trailing non-functional material (sed filtering)?? Did you account for eliminating commentlines (more sed filtering)??

3.) You didn't sort the data - sort is needed

4.) You didn't deduplicate the data - uniq is needed

5.) You can do piping of outputs but I'd use concatenation from other files (you will get MULTIPLE sources is why) - cat needed

6.) To cleanup the raw data when done - rm needed

7.) For more filtering (per #3 above) - awk needed

8.) Data isn't always text from remote sources - unzip/zip needed

9.) To migrate it to etc - mv needed

* Same bs other 'scriptkiddies' like YOU fucked up on in the past vs. myself @ LEAST 5x on since 2010 per my other reply to you.

(You're ALL the same you 'scriptkiddies' - fuckups).

APK

P.S.=> Above ALL else - did YOU write those *NIX commands? You'll also need ~ 14 of them - I don't - HELL I PRACTICALLY WROTE EACH ONE into 1 single multitasking multithreaded PRETTY gui program that isn't interpreted bs & ugly tty term/dos window bs... apk

Re:You miss what other 'scriptkiddies' always do

[barbariccow]

Add "-i" to the grep command to handle case differences. Either trolling or you have absolutely no skill or experience programming, as this is almost completely wrong in every statement.

Wrong: barbicow put himself outta business

[Anonymous Coward]

Wrong: barbicow put himself outta business & EXACTLY how/when/where & why https://linux.slashdot.org/comments.pl?sid=11356847&cid=55558543/

* Unbelievable, lol... same shit I have seen before here as I have said to you before & YOU SCRIPTKIDDIES ALWAYS BLOW IT, every single time!

APK

P.S.=> Do I always have to do YOUR WORK for you? Apparently all you do is use others code in OpenSORES plagiarism (since you don't have the brains to do it yourselves & your prebuilt lego frameworks always turn up bugs too (makes me laugh, the price you pay for plagiarizing boys)), fuckup on tasks WITH TOOLS YOU DIDN'T WRITE YOURSELVES too no less (I don't have THAT problem - THAT is for "noobz" (like you))... apk

Re:Yeah. And?

[Freshly Exhumed]

They're still waiting from 5 years ago for pagefile.sys to populate. "It looked like it was all booted up and ready to go all that time, but in reality the disk thrashing kept going and going forever," complained a researcher.

beowulf

[Anonymous Coward]

Imagine a Beowulf cluster of those!

So?

[Anonymous Coward]

If it works, who cares?

No experience programming? /.ers disagree

[Anonymous Coward]

Doesn't handle all cases I listed (put su/sudo & nice on there too to make a script do all I note). THis does & far more https://it.slashdot.org/comments.pl?sid=11341203&cid=55528501/

* That last part's about my "no experience"...

(Quoted /.ers clearly disagree outnumbering you 10++:1 & so do security pros)

APK

P.S.=> Want more of the same? Ask - I only ask YOU show YOU have done better - that's all - big talk from you? Back it up w/ solid undeniable proof as I do... apk

Missing things still... apk

[Anonymous Coward]

See subject: I don't see a ping command there - speedsup favorites sites you spend most time @ (secures you vs. dns fails + tracking too). Want more? See here -> https://linux.slashdot.org/comments.pl?sid=11356847&cid=55558543/

* You just added the curl (wget works too), right? "Good job" lol!

APK

P.S.=> That's only a way to do part of what my program does in what you put out still (It's a single GUI multithreaded/multitasking .exe, not a batch job via scripting) - Try writing the equivalent of 14++ total batchcommands in script as I did instead of trying to bust my balls - even when a script works & I've annihilated a lot of others try as I proved? You still aren't up to the mark... apk

Still missing things... apk

[Anonymous Coward]

See subject & https://linux.slashdot.org/comments.pl?sid=11356847&cid=55564827/ & nice add of curl (wget works too) on my suggestion + you need ping, sudo/su, nice + more (I always end up having to help you rearchitect your errors - don't feel bad - I did that for years for junior programmers).

* Where's your GUI (I could add argc/argv commandlines to mine but seemed like a risk)?

(I'd also say where's your single part multithreaded/multitasking .exe but I know that's beyond you but even in the world of windows, there's batch compilers (won't build multitasking/multithreading OR AUTOMATED EXECUTIONS I do in my program though - more commands you need would be chronjobs in *NIX) you know (interesting QUESTION - Are there those for you *NIX folks?)).

APK

P.S.=> James Dean (lol) https://www.youtube.com/watch?v=7guzEuV7j9c/ - that's me. what can I say? The Eagles said it best "You were the low down rebel even IF you had no cause - sockhop, sodapop, basketball & autoshop - the ONLY THING that got you off was BREAKIN' ALL THE RULES!"... apk

Wrong - no linux

[Anonymous Coward]

Real optimized supercomputers dont run Linux. For instance, IBM Blue Gene(?) runs Linux. But it turns out that Linux is only used for I/O, to distribute the workloads out to each node. Then a special minimal kernel runs the computations. So Blue Gene is listed as a Linux cluster, but that is not true. If you can squeeze out 3% more performance by modifying the kernel, you will do that. And Linux is open source and easy to modify, so of course they choose Linux. Unix is often closed source and more complex code.

Linux runs fine on large HPC clusters (such as SGI UV3000), but some workloads can not be run on clusters, typically business workloads (SAP, databases, etc). These workloads can only be run on a single large server (scale-up instead of scale-out cluster). These huge business servers weigh 1000 kg, have 16-cpus or even 32-cpus. One SPARC Fujitsu M10-4S even has 64-cpus. Mainframes belong to this category, large business servers. SAP HANA is for analytics and analytics is easily parallellisable fit for number crunching HPC clusters. Business workloads use lot of syncronizing between cores, and if you have too many cores, synchronizing will kill performance. So business servers are typically 16-32 cpus. If you see anything larger than that, for instnace, 100s of cpus, then it is a cluster. And clusters are not synchronizing heaviliy all the time, they run independent workloads, and at the end they synch. Clusters are run by one user, who starts up a workload that runs for days. Business servers are used by 1000 users simultaneosly, all doing different things: accounting, pay salaries, database, etc - which need heavy synching.

Large scale-up business servers are exclusively Unix and RISC (and mainframes). They cost many millions. One IBM P595 that was used for the old TPC-C record, had 32-cpus and costed $35 million. No typo. You could buy many Linux clusters for that sum. Clusters are essentially a bunch of PCs on a fast switch - so a large cluster is cheap. A scale-up business server is advanced and built for scaling, and therefore costs much more than a cluster. The cost increases exponentially with the number of cpus. Only the last year, the first large Linux scale-up server arrived. It has 32-sockets, and I expect it to have bad perofmance as it takes decades to scale well.

Understand: Barbaricow = off topic troll

[Anonymous Coward]

See subject: Barbaricow came in here totally offtopic to try "get the better of me" only to collapse vs. facts I put out https://linux.slashdot.org/comments.pl?sid=11356847&cid=55565189/ & others like it where I had to correct his errors (he thought grep & some switchwork on it did everything my program does).

His 1st post here in this exchange proves it. The rest of his posts were EASY to blow away also.

He also tried "putting words in my mouth" I never said which I corrected him on also here https://linux.slashdot.org/comments.pl?sid=11356847&cid=55558139/

APK

P.S.=> Try understand that well over 1/2 of the stooges giving me guff are either webmasters OR advertisers, malware makers/botnet herders, or inferior competitors with numerous sockpuppets - doesn't matter - I swat them down with facts they can't overcome... period (& it shows every single time they try it & FAIL, lol)... apk

Re:Yeah. And?

[iggymanz]

They got through that but now they're waiting for them to shut down, the 2011 batch of windows supercomputers is currently displaying "Installing patching 354 out of 1023 Do Not Power Off"