Slashdot Mirror
IBM Exec Says no Large Web Servers on Linux
Accidental
Angel writes "As reported in
InfoWorld, Tony
Occleshaw, IBM's software strategist for Europe, Middle
East, and Africa said at CeBIT today that
"No
one runs large, million-hits-per-day Web sites on Linux." "
Well, we served 640,000 pages on Wed on this Linux box. And the
server load is only 2.00-3.00. I figure this box can handle
around a million. The
adfu server (also Linux) did around a million hits total that day,
if you combine banner ads + layer HTML).
IBM is probably just saying what they 'think' which is different from what some of us 'know'.
I think IBM just needs to give Rob a big bucket of money to act as a consultant and show them how to set up linux servers.
But this site is always one of the slowest I visit. I know you get a lot of traffic, but the ad server's performance means I have to hit stop to see the meat of the page probably 10 to 20 % of the time.
Also, the first time I went to this comments page, it was blank. The comments pages are regularly f*cked up.
Seriously, Rob, Slashdot is the poster child for IBM's point on this....
For the price of one decent AIX box you can get
at least four twin Pentium Linux boxes. When it
comes to betting the farm on a platform, I'll take
a good base and three live spares any day.
I like slashdot, but this place bogs down (or worse) whenever there is heavy traffic. And people are mostly just making little posts or reading what other people said.
If I were spending money here and gave a credit card number - and then read that there was some server problem (not uncommon here) I'd be really fricken pissed not knowing if my transaction went through or not, or if it went to some limbo only to go through at some later time (I may have to call you or my credit card company to find out what actually happened - which is what I'm trying to avoid if I'm doing business on the net), and I'd probably wouldn't return to do business.
I don't know if the problem is Linux or your hardware or whatever, but what you're running is peanuts compared to what I'd like if I were running a serious ecommerce site.
really
this site has always been fast loading for me
maybe you should find a new isp . . .
IBM is talking out of its ass on this one. I've configurred, setup, run, and administered a linux web server that performed over a million transactions in a day with the same fingers that typed this message. Admittedly, this wasn't without some tweaking (Use The Source, Luke..), but it's quite possible to do it.
Given that the machine's gone through 2 hardware upgrades since I left that employer, I can only assume it's doing even more transactions in a day.
In all fairness, AIX is a nice operating system in several regards. Its disk susbsystem and LVM are the best on the market hands-down. (Ever reshuffled 500GB of file systems on a production database server during peak hours without anyone having a clue it happened?) Its parallel and SMP environments are pretty slick and easy to administer.
IBM's only real problem with AIX and its RS/6000 line is that it loads a good hardware & o/s pair down with crap/bloat/segfault/waste/coredumpware like Lotus, Xcom, and all the other silly "standard" software offerings that chew on resources Just Because They Can.
IMHO, IBM needs to be a little more honest in how it presents things. (Or maybe the execs need a few blows from the Clue Hammer[tm].)
hum...
:))
Yes, we do not do 1 million hits a day, but I have several high traffic web server running linux. The daily average hits on the machine is only 861136, it is not a million... but the load on the machine is nothing (p2-450, 512Mb, load 0.07 !!!), using as the file system a NetApp for the webpages, the kernel of the machine 2.0.36
[stephane@mall0 stephane]$ uptime
12:27pm up 72 days, 19:42, 1 user, load average: 0.08, 0.02, 0.01
Of course you have to know LINUX, I am running LINUX for 6 or 7 years now. I had some problems of two many files open, but this problem have been solved by hacking into the source code, but now with the new kernel 2.2.x, it is no more necessary, the limits is 4 times higher and can be changed on the fly.
I am sure, this machine and the others we have can easely handle more than 1 millions of hits a day without a problem, the problem will more be bandwith and CPU power than LINUX
We are also using LINUX for a high traffic Oracle Server, on a dual P2-450, 1Gb of ram, LINUX kernel 2.2.2, and it run smoothly, without any problems.
I think, IBM is afraid of LINUX as a potential competitor for AIX.
The main problem we have for LINUX, is not with the system, the problem is the Hardware, it is still the lack of scalability, on a SUN enterprise 4500 for example in case of a motherboard failure you can change it on the fly, or add memory, or do some modification on the system without having to reboot the machine, it is not yet possible on LINUX, only because of no hardware like that available. The Day this kind of hardware will be supported by LINUX, then, the other systems might really be scared !
stephane@younix.com
all the hardware in the world bogs down with .5Mhits/day on 1.5MBits/s
The very first dilbertism
I realize it's not Linux, but FreeBSD is being used is some very large sites, such as yahoo.com, cdrom.com, etc. If the point of all of this is that you HAVE to go with a commercial/overpriced OS for serving that kind of volume, those two sites are good examples to the contrary.
FreeBSD is really able to deal with millions of hits/day. However maintaining a FreeBSD system is a real pain in the ass. IMHO if FreeBSD could have a decent distribution (as RedHat or Debian) it would be *a lot* more interesting than (the hyped) Linux. I've seen some posts regarding a Debian/FreeBSD in the Debian devel mailing list some weeks ago, but I guess it won't go too far, since that guys appear to be more concerned with "Linux" instead the distribution itself. So, let's expect that the few people really working in the linux kernel can fix it to make it have more scalabitily. Or earn more money to buy a Sun workstation...
Could you build a slashdot type system that could handle millions of -comments- per day? There's probably a factor of about ten between the number of hits a server can handle and the number of real transactions (source: top of my head)
yes, Linux is a potential competitor for AIX, and when it becomes a real competitor, expect IBM to move the two closer together -- IBM is in the solutions market, and if it can sell Linux solutions for less with the same reliability, then it will.
That said, when IBM sells a system, it can take no chances -- IBM is the type of vendor that needs to be able to give GARUNTEED uptime for a given system, and Linux at current is too fickle about its setup (p.s. how many of you -can- make a flaky Linux installation if you want to? I can --
compile the kernel with -O6, agressive strength-reduction etc.)
If you want something really huge, like Amazon.com, or hotmail.com, the first thing you do is ensure you use multiple machines with failover.
And for this Linux is very cost effective, since it runs very well on commodity hardware. If I could choose between 20 x86 based servers, or 10 Sun servers with twice the performance (regardless of price), I'd choose the 20 x86 based ones, simply because it is better to loose 5% of capacity now and then, than loosing 10% of capacity a bit more seldom.
Besides, Linux has been demonstrated on 16 CPU ultrasparc machine... Doesn't scale, you say?
As for Hotmail, last I heard they ran Solaris.
And as for Slashdot, judging Linux' reliability based on one site, is hardly very valid. It is also even less valid considering the lack of failover, etc. that any big e-commerce site or similar would have.
In the prime time (about 7 - 9 pm) slashdot slow to the point of being unusable. So much for scalability.
That's my server setup.
So it all boils down to what "a hit" is.... Is it 8kb of static data, or is it 200kb of dynamically generated content, or something in between.
Linux 2.2 is much more scalable than 2.0, and I think in the future it will get even better. However Linux's SMP is optimized for 4 CPU's or less. When you try running it on a box with more than 4 CPU's, it isn't nearly as efficient as it could be. On the other hand, this is exactly what commercial UNIX's are designed to do, and they do it better than anything else on the market.
Think PowerPC Pad!
Isn't Oracle 8 for Linux free only for non-commercial opperations? Now that slashdot is a commercial site, they've got to be careful not to use software labeled "free for non-commercial use."
There are a great number of web-sites that run, very successfully, multi-million hist/day sites using Linux.
The JPL Pathfinder site ran under Linux, and received some of the largest number of hits during that time. The IMDB, as far as I know, uses Linux servers, and so does TuCows - all successfully reaching huge hit-rates.
MeThinks mr.Occleshaw needs to eat his words.
HarryZ
It's fine for me. Maybe it's time for you to look into a new isp.
Well, what would you bet on?
IBM 'Why did that partition go stale' AIX?
Sun 'Hmmm, there seems to be a bug in the network driver' Solaris?
HP 'Where did that system crash dump go' UX?
Or maybe 20 Windows 'Ok, send Trained Reboot Monkey 2 to server 5; it's bluescreened again' NT boxes?
Face it, if you've ever been involved in any sort of major systems you know they *all* exhibit some problems during heavy load increases. Some more than others. That's why you have sysadmins who can fix this stuff when it breaks. And it's why you have trained monkeys to reboot the NT systems.
If Slashdot were an e-commerce site they'd have both the sysadmins and redundant systems, they'd make certain they never had a load above 1 and, if possible, keep one or more cache servers outside to keep the load lower on the main systems. Linux is at least on par with the commercial vendors when it comes to stability.
There is a significant difference between running on a 16 way machine, and running 16 times faster on a 16 way machine.
If you run a commercial UNIX, you can expect close to linear scaling with multiple processors as everything is done very fine grain.
This is not true of Linux, which still doesn't do SMP particularly well as everything is still to course-grained.
When the average Linux user has a 16 way box, this will improve, but currently most Linux users have uniprocessor machines and it is primarily a uniprocessor OS.
Who could read millions of comments in a day? :)
/.'s not coded well. Just my $0.02.
Well the comment that ./ currently serves well over half a million hits a day kind of puts pay to the web server issue. Indeed many other servers appear to surcum to the "slashdot effect"
As for the lack of applications moan what is stopping IBM from attempting to create the "Holy Grail" of applications. That is an application which runs quickly and does the tasks the user needs. (N.B. this also relates to the MS Offioe on Linux thread.)
I am currently working for an ISP that is taking advantage of affordable price of Linux to create a web farm which maintains high availablity for all of the round-robin IP addresses via Fake. The Fake package combined with Rsync make it trival to produce a scale-able web server enviroment.
Btw, we did evaluate AIX & HACMP as a possible solution. The turn-around time for leasing additional RS/6000 and the design of HACMP did not lend itself well on being able to grow a web farm on the fly as needed. The reality was that HACMP appeared to be much more difficult to administrate and *scale* than using the Linux Fake package.
Also, IBM is right that no one runs *1*-million-hits-per-day web sites on Linux. That is too trival. We are gearing to be prepair for 3-5 million-hits-per-day. We saw a web server become unavailable shortly after a popular prime time news program announced the address. With Linux, that will never happen to us.
How much traffic do you think these guys handle?
--------------*
Welcome to wcarchive - home FTP site for Walnut Creek CDROM. There are currently 3600 users out of 3600 possible.
This machine is a P6/200 with 1GB of memory & 1/2 terabyte of RAID 5. The operating system is FreeBSD. Should you wish to get your own copy of
FreeBSD, see the pub/FreeBSD directory or visit http://www.freebsd.org for more information. FreeBSD on CDROM can be ordered using the WEB at
http://www.cdrom.com/titles/os/freebsd.htm or by sending email to orders@cdrom.com.
"really
."
this site has always been fast loading for me
maybe you should find a new isp . .
Dear Dick head,
I've experienced the same problems, and it has abosolutely nothing to do with the isp. The problem is that the banners take a while to load which is *extremely* annoying since it means that you have to press stop to see the content.
At my work we have almost 100% unix desktops.. We have a few NT Terminal Servers + Metaframe for the few windows apps that a few users need.
Note that "scalablity" does not equate to "ability to support SMP". A shared memory SMP system will only make a difference to *CPU Intensive* tasks. A web server's main job is shuffing data between disks, memory & NIC(s). Further when it comes to operating an SMP there are issues of cache consistancy, bus bandwidth, memory management and program design. If you want to do this with Intel processors than Sequent is probably a better place for hardware design inspiration than IMB
every other site works fine except for slashdot.
Oh, and I don't live that far away -- only 13 hops.
All I can guss is that your ISP sucks. I'm sitting on a fractional T3 here and Slashdot's loads at a typical 10-30Mb/s which is better then most sites..
Well Cisco, a big advocate of e-commerce, does something like $30,000,000 per day (check their site, they should be loud about this). However they probablt aren't using a Dual P-II, and most of their products cost more than $30 :-)
In order to get a multi processor machine who's perfoemance is anything like a linear scale with the number of processors you need
a) High capacity memory and I/O buses.
b) a sophisticated MMU.
c) complex cache management.
It also helps if you have applications which are specifically written to both operate in an SMP environment, written to "expect" the "right" number of processors and that the OS can easily recognise when a process group works best being scheduled together.
AIX may have LVM and JFS, but don't try to run the latest X programs on it because they're still in X11R4 or X11R5. Its various system utility programs (sed, grep, awk, etc) are outdated and don't support nearly as many features as their GNU equivalents. Their default window manager, CDE, is not threaded and one operation can lock the entire window manager up while it runs. SMIT is ugly and awkward to use. And just to make matters worse, everything is *Motif* based and vomitous in appearence.
Take an AIX box and install bash, perl, gnu (sed, awk, grep, tar, etc) Xfree86, GTK+, Lib (jpeg, tiff, png, [un]gif), Enlightenment and Gnome and you would have a decent system that coincidentally would be 95% Linux.
I think Linux would perform as better as FreeBSD on cdrom.com.
Cdrom.com is serving mostly static files via ftp, and given the amount of RAM/bandwidth/disk the cdrom.com box has i don't see any problem on Linux doing the same job.
On the other hand, i would give more credit to slashdot, who also get's thousand's of hit's/day ( not as much as cdrom.com ) but it is servig mostly dinamyc content rather than static files and also is running on a single box.
Another, same hardware config, gets 800k a day, and its pages make heavy use of SSI "cmd=" include directives--seven per page, so there is quite a bit of forking and execing going on.
Load average on the first server is usually in the 2 to 4 range, with spikes up to 8. Interactive response stays great, making it a pleasure to do administrative tasks on.
Based on the data from Alexa and 100hot.com, our servers are at least as busy as anything IBM is running.
(I'm posting this anonymously because we have a web-based business, and details on the number of hits would be valuable information for our competitors)
Out here in the Antipodes there is a noticeable slow down on the internet from 5pm to 9pm. This is best explained by the fact that there is only one line from Australia to the rest of the world. And seing as most of the population is within close proximity to the same timezone this poor little line gets overloaded. In fact during other times slashdot is quite fast over the modem line.
Slashdot gets 640,000 hits per day and you *think* it could handle a million? When building a million-hit per day website - you need to plan for the 5 million hits/day mark or more. You need to worry about peaks more than averages - downtime sucks.
And bear in mind, Slashdot isn't a very high-tech website. No advanced server-side queries, no encrypted pages, limited databases, no dynamically generated pdf files, etc. If you're building a storefront - you WANT all of those things.
Don't get me wrong, Slashdot is a nice site but it doesn't do everything a commerce site would do and lacks the reliability demanded by such a site.
Most of this may not be a Linux issue. Commodity hardware just isn't there in many respects. Someone mentioned 16 way SMP boxes for instance. Sorry, you just *don't* know SMP then - SMP doesn't quite scale nearly that well.
And whoever said this guy was an RS/6000 rep? Aren't RS/6000 geared more towards technical/scientific computing? He could be a AS/400 software specialist and believe me, those boxes are far more scalable than a commodity PC. He could also be a mainframe guy too (where a million hits per day is peanuts).
The page is very complicated to parse with a lot /. could serve 2 million.
of tables and images etc. This takes time too.
Another thing is that the server has no flat
pages...ie all the pages are served while connecting
to a database and running cgi written in perl.
These all combine to put a lot of load. Serving
flat html pages with no image content and very
few tables
Most ecommerce sites are not as complex as Rob's
site and very few add as much functionality
along the way.
.. since there is no other Linux website that gets as many hits as www.microsoft.com .
In the company that I work for although most of the managers and system administrators use Windows NT the customer service reps use HP-UX workstations. They workstations are just a tool to get things done, ie nothing really cool about it.
Another UNIX workstation center
How about a Linux shop like dejanews.com? They get tons of traffic. Of course that is spread accross their 17 web servers and who knows how many other backend systems. The real solution is to use some load-balancing device. Round-robin DNS just doesn't cut it anymore.
FreeBSD is extremely stable under heavy loads. Sending gobs and gobs of data out every second is a high load. Plus the most important part for the commercial market is FreeBSD maintainers don't put last minute "features" into their releases.
Another nice feature is FreeBSD will compile 99.9% of the source which compiles on linux and can run probably 95%+ of linux binaries through their linux emulator.
Hey guys
s t=www.dejanews.com
http://www.netcraft.com/cgi-bin/Survey/whats?ho
Isn't DEJANEWS your idea of a 'big site'?
Some guys just need a few slaps
-Yist
When I use Commun4.08 on Linux, it is kind of slow. KDE Konqueror makes it blazing quick.
host -l dejanews.com | grep www.dejanews.com
Shows all 17 of their webservers. Who knows what all those other systems are for.
Please, dont say something as stupid as
"FreeBSD is easy to maintain".
It isn't, even if you think so.
Debian, for example, is several orders of magnitude more sophisticated and easy than FreeBSD. It's unlikely that Debian or RedHat will have a BSD version, so it's unlikely that FreeBSD will have a decent package management system in the near future. Even so, its kernel is really cool.
And thats what makes it better. No cut-and-paste from here or there to get the system working. Especially no mangled login messages in /var/log/messages after upgrading the kernel (anyone know how to fix that in redhat 5.2?? I've got two systems doing that to me). The CVS tree is really nice too.
man, if you like to rebuild everything with make world, and keep your system updated with cvsup, and using /usr/ports as a can of trash for everything your poor package management can't maintain (shouldn't it be /usr/local? cant't be, freebsd uses for usual packages!) you're sick. what's next? that msdos systems are 'pieces of cake' to maintain?
Nice try? Any of those 13 routers (if fixed routing is used) can cause lagging. Have you tested all those machines in between?
Well, the place I work for's primary virtual host server used to be a P120 ( running BSDi 2.1 ). We used to regularily handle over a million hits per day, mostly on one site. It only bogged down when someone hit a large cgi program on one of the other 500 or so sites..
.03 :) the main hit on a webserver isn't CPU, unless it's main job is cgi, it's mostly I/O. Big disk caches, fast seeking disks, and low-latency network connections are more important, as is a fast context-switching architecture when things get really busy.
We've got one P2 running FreeBSD & Apache which regularily deals with 45GB per week, one week went over 90GB and about 800k/hits day.. I'd still call this a small server, but it's traffic has got to the point where small kernel tweaks make a big difference. It's still got lots to spare, though.. usual load avg is about
IBM is _always_ on it's own side only, isn't that obvious??? It's business folks.
so just installing this software, that also runs on linux, makes aix be same as 95% linux, heh? then i don't use aix running as operating system on a server. if i did it would not be compiling X programs, or running X binaries...it is buggy!
I have to use Netscape on '95 at work where it can take a *long* time to render the page. At home, I use Lynx on Linux. The page loads and displays *far* faster at home over my 33.6 (which, due to the ^%&*% phone company, can only do about 26k) And, yes, to make the comparision more fair, that is even with images and java/javascript turned off at work.
I haven't tried IE. I have to use Netscape for other tasks (it is the campus standard) and I don't want it GPF'ing after 5 minutes (If I have both open, after 5 minutes netscape will GPF. Re-boot, open both and get it again, ad nauseum.)
I agree with the view that Linux is a good OS for small or dedicated servers (much better than NT), but it lags behind UNIX operating systems like AIX, Solaris, HP-UX and Tru64 (OSF/1) in terms of features necessary to power mainframe-replacement systems.
As an example, the best SpecWeb96 result was generated by an IBM mainframe running OS/360, but the very-close second spot is occupied by an IBM RS/6000 running AIX. All the other contenders tend to run commercial UNIXes (I don't think anyone's even bothered to submit Linux results).
Of course, most of the advanced UNIX operating systems don't run on PCs, and the very advanced features that provide such good performance on `big iron' make them bigger, and slower than simpler operating systems when run on cheap hardware.
Slashdot does not seem to perform very well, at least to me.
Funny, /. seems slow to me, but I have always been impressed with the speed of all the Yahoo! sites. Perhaps it's down to the various hops in between.
... seen both - fortunately the AIX was a RAID 1 system and the Solaris had two ethernet cards. :)
One weird plus for AIX is the "verbose" error messages. They don't help you solve the problem, but at least they give you something to parrot back to IBM.
- Not really an AC - only broken post code in KFM.
> FreeBSD happened to be better able to handle
> brutal loads when Yahoo was setting up shop.
Yes it was. More to the point, it still is, and will likely remain so for the foreseeable future.
Who cares if the page comes off databases with perl? I'm sure the front page is cached in ram. At the worst case, only the first viewer that sees a new page would feel the sting of the db hits.
/. is badly coded, with those stupid green boxes with black highlights that nobody can see anyway. I wouldn't expect any better from ZDNet, but a hardcore hack site like /. should be times roman on battleship grey. It should look like a geek site, and be hyper-fast loading too. We come here for the tech info, not for the dumb Einstein faces.
/. use to be much worse, but it is still an EMBARRASSINGLY slow load, and I've seen it on Macs and NTs over modems and T1s, with all different ISPs too. The fact is that
I'm not against graphics but it should be limited to photos that convey content, not eyecandy. Heads don't need ugly green boxes; large text in a different font or color is just dandy.
scripting.com is an example of a well-designed site running off a slow server. You can tell the difference.
I saw many so called "problem weeks" on AIX and ;>
Sparc boxes too. So that is not an excuse. Any system needs expert guys to start working properly. And what does IBM propose for large scale web serving ? NT ? *laughs* Given the experience i had with NT on production servers
i'd use NT for large scale disasters
Actually it sounds from his title like he's in the software division, which has nothing to do with the hardware sales (including the OS).
Read my lips: Don't Load the Banners.I use Lynx as my web browser, so of course the graphics don't get loaded.I get the content I want from Slashdot *very* quickly, always, and I don't mean gross throughput, I mean latency.As soon as I hit a key for the next heap of information, *WHAM* -- it starts arriving with no perceived delay, which would put it below that magical 200ms threshold.There is nothing wrong with slashdot's server; you just need to fix your broken browser configuration.
-- Guges --
Pentium 100, 128MB of EDO DRAM, tulip 10bT, Apache 1.3.4 on kernel 2.0.36: 1.2M hits/day peak, 0.9M hits/day weekday average. Most pages are shtml too.
Also make sure that the new ISP isn't using NT.
That's a sure way to kill your transfer rate...
I noticed the adfu.blockstackers.com's slowness for about a day, when I
;-)
first started reading Slashdot. But ever since they changed their address
to 127.0.0.1, they have been lightning fast, even if a bit boring in
terms of content.
Wow, I just noticed that a *lot* of businesses are using that 127.0.0.1
address. That machine must be one hell of a server.
Tony Occleshaw is a first line manager (meaning, one step from the bottom of the food chain) and has less people working for him than I can count on one hand, even after that nasty accident in shop class.
I've done some checking around on him. He runs his mouth a lot to the magazines, claiming these grand titles (which are untrue), and basically making public statements that go against the strategies of IBM.
Next time Occleshaw runs his mouth, I'll post the names, e-mail addresses, and phone numbers of everyone in his chain of command on up to Gerstner so you guys can let IBM know what you think of him.
My personal beef with linux has been (for some time now) File Descriptor limitations... that was with 2.0. I haven't played with 2.2 in a large enough production environmen to see if that's still the case. (I'm told that 2.2 handles FDs in a far more elegant way... good!)
Umm, tables have nothing to do with the server man!
/. can run on a 486 if it was made in C and not perl
/. in terms of hits etc, has got to be stupid to do it in perl, perl is slow assed, just as bad as java.
And 2) delete perl ROB, go an learn some C, like real men do!, C can be as easy as perl if you are a good coder, and perl can be worse if not harder to read and more complex to do shit
I bet
2million pages is only 23 per second which isnt too hard to do, but that is pages, not connections, so each page coul have 3 gifs, so thats 23*4 serves / second, so?? Make a ram disk put the most scommon data in ram.
Anyone considering a large site larger that
C all the way man
2 seconds on my IE5
IBM does the olypics, it got more than 40 million
hits in a dat, and may have peeked at 20000/second and one time I believe
Could linux do that?
It all depends how much $$ you have for what ever task you have, but linux isnt the Jesus christs of solutions. IBM with $50m can do better
I mean a box with gigBit Ethernet and 16 450mhz PowerPC cpu's will kick but then again if linux could run on it...
Big deal, its a vender solution, and you dont really know till you try/
Do you have any idea how many systems excite.com uses? just visit their office and you will see rack after rack of 30 computers + sun boxes
But my personal webserver, is piddly, running 3000hits day
SO WHICH SITE?
why don't you give IBM an example?
However, that's not how the comment was framed.
When the task will make NT shriek, jump up out of it's NT box, and run away screaming into the night: THAT's when you need IBM.
as far as i know System/390 starts from half a million up.
www.dejanews.com runs on 17 servers...
I'm really surprised no one has brought this one up...
While everyone's been talking about OS/Hardware issues, what about the webserver? I'm sure that to most people here (as well as anywhere else) apache is what you assume without asking. But
most popular != fastest, and apache is nowhere near a speed demon (well, maybe if you're comaring with IIS). There are certainly a bunch of alternatives that push the envelope a *LOT* farther than you can with apache on the same iron. Things like (in my WAG estimation of increasing speed):
1. Roxen (open source, free, multithreaded with built-in database hooks, fastcgi)
2. AOLServer (closed, costs money, multithreaded with built-in database hooks)
3. Zeus (closed, costs lots of money, multithreaded with built-in database hooks)
Tens of millions of hits should be possible with any of these on properly specced intel iron under linux.
Of course, I'm going to get flamed to a crisp for even suggesting that people contemplate the thought of spending money on any software...
How does OpenBSD (x86) performance compare to FreeBSD? Does glide work under either/both?
> what's next? that msdos systems are 'pieces
> of cake' to maintain?
What the hell? They most certainly are.
Sound like FreeBSD isn't any harder to maintain than Slackware, though. So, should I go for FreeBSD or OpenBSD?
Ok, I've been wondering, so what's so cool about Roxen Challenger?
If it's only OS that is a 'problem' with 1M hits per day... If you have a sysadmin that is a moron, and doesn't have good hardware... You're dead.
;) is that amazon.com is having only 1 machine/ip 'assigned', while barnesnoble.com (NT) has 2 :) "Just in case..." ;)
What Netcraft.com says...
www.register.com - Linux/Apache
www.genocide2600.com - Linux/Apache (believe me, these guys are getting a *LOT* of hits per day)
www.quotes.com - Linux/Apache
www.necx.com - Linux/Apache (big online computer shop)
LinuxMall.com (which is getting a *lot* of hits too)
etc, etc...
But what makes me feel good (when it comes to NT/UNIX war
hrm....
/. :P
I'm in australia, and no problems with Slashdot loading here.
so um.. get a clue.. its not
smp != meaningful measure
SMP is cool, and I sure would like to have eight P3's in my box.
But it's way overrated. I've coded for various parallel architectures, and shared memory sucks the most. Shared memory multiprocessing (e.g. SMP) is a dead dog for parallel design scalability. You can go 2 or 4 or 16 but not 128 ot 512 processors. The machine that beat Kasparov in chess was a 32 node SP2 by IBM, and, while it could have been done as SMP, it would have cost zillions
SMP should be called "small machine parallelism"; A farm of n CPUs will be cheaper than an n-CPU system for suitably large values of n, for the right class of problem.
Ummm, I'm going to make a wild-assed guess and say it's:
http://www.scifisites.com/
How I came to that conclusion, I really can't put my finger on. Maybe it was the URL in the message...
John Waalkes
jwaalkes@edge.net
Two words: Internet brokers.
Forget about C! Real men learn Assembly Language. All serious server side apps need the power and speed of Assembly.
/. could run off an Amiga 500 if it was made in 68K and not C.
I bet
Yeah, somehow I have to think that SABRE, the reservation system Travelocity runs on, can do more than that in an hour.
Our websevers here at work serve 8 million a day between the 3 of them...they are 233 Pentium machines with 128M RAM each...that article is total BS
This is traffic from midnight to 3am. Imagine the traffic during the day (5-10 times more). These machines barely break a sweat during the day.
_ _______________
Server Version: Apache/1.3.3 (Unix)
Server Built: Oct 29 1998 12:35:56
_________________________________________________
Current Time: Sunday, 21-Mar-1999 03:11:03 CST
Restart Time: Saturday, 20-Mar-1999 23:30:27 CST
Server uptime: 3 hours 40 minutes 36 seconds
Total accesses: 49882 - Total Traffic: 374.4 MB
CPU Usage: u104.1 s138.4 cu272.53 cs67.78 - 4.4% CPU load
3.77 requests/sec - 29.0 kB/second - 7.7 kB/request
28 requests currently being processed, 37 idle servers
see above comment:
5 8230&pid=0#2221
http://slashdot.org/comments.pl?sid=99/03/20/11
This guy is just some media whore looking for attention! Find out who his boss is and get him canned!
http://www.netcraft.com/cgi-bin/Survey/whats?host= www.scifisites.com
netcraft says IRIX, not Linux...
maybe that's not his company's site...
It is actually free. You can download it from www.aolserver.com
The 2.0 kernel would have trouble with that because its open() call was so slow and Apache does no internal caching. But the dcache in 2.2 should have resolved those problems. (I assume from the fact that Slashdot is now noticably speedier that it did :-).
-- Eric
For example: Database applications, AIX supports large memory systems (3+GB), journalling, raw i/o, etc. Assuming equivalent hardware, I'd estimate AIX would be able to get 50% better performance over Linux running a typical database task.
They're both great. Linux and FreeBSD are fast,
robust, and free.
No, Linux is now better under heavy load. It is because I say it is, just like you.
If you checkout www.webring.org/cgi-bin/wrstats you will see that their peak daily hits for the past 8 weeks was 9,535,049--which is way more than a million.
Cheers
I've done a little bit (just a bit more than your playing around, not much) and I noticed it running noticeably slower than mySQL. I think this owes to the architecture it uses. It spawns a new process for each connection (ala Apache, except that it isn't just a simple fork, it starts an entirely different program from the one that listens for connections.) Since my application was making lots of connections for small transactions (put this into the db. new connection. look this up. new connection. etc.) it went pretty slow.
On the other hand it seemed quite stable, had good docs (better than mySQL) and seemed like a solid piece of work.
YMMV
But is webring a single box or multiple servers?
There's probably a factor of about ten between the number of hits a server can handle and the number of real transactions (source: top of my head)
Been there, done that. Using a P100 with 64megs, I could get 66 pages per second, using nothing but static HTML. Replace that HTML with a mod_perl script that changes out the URL's randomly, and you suddenly drop to 20-25 pages per second (assuming 10 URL replacements/page). These measurements were of the time when CPU hit 100%.
That doesn't even go into the overhead of db transactions for the backend.
Why not just use telnet to browse the web instead of using that memory-hog lynx? Oh I dunno.. maybe because some of us would like to live in the 1990's and actually see graphics instead of just black and white text? Grow up luser.
after all, a strong point! :)
it's really easy to maintain a system using
'make install' and maintaining a database of
what is installed. so you just need to type
some rm|cat|grep pipes to remove a package.
also, this system is perfect: you always know
that a package will work with the stuff installed
in your system, right?
and when you need to build a system that is a copy of another, you just need to 'make install' everything again. cool.
freebsd management is a piece of cake...
I agree..
So, I'm likely to love FreeBSD.
RedHat is a plague in the Linux world.
If you use it to judge 'Linux' you're likely to stay with *BSD.
But have you ever run Debian?
Hrmm...Got to digging arround on netcraft.com and found a Microsoft Site is running an "Undisclosed" server on top of linux.
It's supposed to be testing some new interoperability products and research material.
Check it out.
http://eggplant.rte.microsoft.com/
-Drat
Oopsie,
http://egg.microsoft.com is the Linux server...
not eggplant.
-Drat
Well put - I've had the experience of working at a site with NT (ugh!!!), HP-UX, AIX ("Aint unIX), and Solaris boxes, and each description hits the nail right on the head (been there, done that, seen that...:-)). Even the proprietary "commerically accepted" OSes have their own quirks. If there was such a thing as "the perfect OS" out there all of these discussions going on over the past 40-50 years would be moot...:-|
There was nothing special about the hardware used for the Olympics site.
It is the software that made it special.
The olympics web site was made up of a farm of many geographically seperated web servers. One master cluster, representing the whole of the Nagano web site, made up of smaller clusters geographically dispersed.
Whole web servers can and did go down, and nobody noticed.
This technology is coming to fruition under Linux under several GPL'd projects.
Once this technology becomes a little more mature under Linux, I think it would be a tremendous boost to corporate acceptance of Linux if we did the same thing.
IBM will no longer be involved with the Olympics after the summer games in 2000 so there is an opportunity here.
Where are you, Linux International? This would be a great opportunity for all of your business partners to make some big bucks off of Linux!
Hell yes I like to rebuild my entire system by typing 'make world'. What could be easier? Download 5-50 new .deb or .rpm's? You've got to be kidding.
/usr/ports a 'can of trash'. It's nothing but skeleton files, a makefile which:
I would hardly call
* Fetches the source
* Patches the source
* Builds the source
* Installs the binary
* Registers the installation.
It's far less likely to blow up in your face than the rpm database. Currently I've got skeleton ports for over 2000 separate applications which are almost guaranteed to compile, I update this collection almost nightly. The entire 'ports' directory takes up a little of 100 megs.
You can call it trash if you want to, sounds to me though as if you've never really tried it. I myself love it and wouldn't trade it for the world.
And finally, to answer your last question, Yes, MSDOS is easy. If you've got problems with DOS then you probably should reconsider running Linux.
I'm crossing my fingers that their upcoming port of the Real G2 player will be up to par in functionality with it's Windows(tm) counterpart.
Come on REAL!
A serious business won't rely on an operating system where you *CAN'T* tune the kernel that way.
No a serious business won't rely on a system were you HAVE to tune the kernel.
AIX tunes itself. Let me see if you can understand this. Lets say the "maximum XXX" is to low. AIX automatically detects this and retunes the kernel on-the-fly. No need for reboot, no need to "recompile".
I have worked with severel Unixes in the past where you could/had to recompile the kernel. I don't want back to that.
AIX 4-ever.
I just want to say that you should use to tool best suited for the task.
Of course you could use Toyotas to haul timber instead of a freighttrain. Look ! My armada of 200 Toyotas with extra fueltanks and kevlarwheels can haul just as much timber as your multimillion dollar freighttrain. It doesn't matter that they break down now and again because we only buy new ones ! It's still better price/performance ! Oh, yeah you have to recompile the engines...
Remember, diversity is good, diversity breeds good OSs.
Why don't you verify before you post? Bummer...
It's amusing to me how many people here are reacting so drastically to an IBM employee's statement of a fairly reasonable position. Let's get some perspective here -- IBM just endorsed Linus extremely recently. They're not about to tell people to stop buying high-end AIX, etc., boxen because Linux does everything.
This isn't just because they're trying to protect high-end sales. It's also true that (1) Linux isn't as robust an OS as the high-end UNIX platforms, and (2) Linux hasn't been in high-end production anywhere near long enough, or widely enough, for it to be a safe option for corporate IT.
Callup can be fun, eh?
;)
Agreed.. the whole bad thing about Linux is the distribution thing. There's one clean multipurpose distro of FreeBSD (it's called FreeBSD strangely enough), which provides options for different types on install. The setup 'GUI' (text-mode) is much simpler.
I just left a firm where I had to use SuSE, supposedly hyped as 'not too bad' in the grand scheme of distributions. The fact that the setup tool is 'Yet Another Setup Tool (YaST)' filled me with dread. Kernel config? Well, it's got a nice X11 interface, but it's still a hack job. FreeBSD has simple kernel configuration files (which may be generated by GUI front ends, or text editors).
Packages/Ports... they thrash RPMs and are easier to use IMHO, since RPMs suck so much.
I started with Linux (Slackware) about 4.5 years ago, made the transition to FreeBSD, then had to go back to Linux. Every time I use Linux (whatever distro) I just wish I was using FreeBSD instead.
If slashdot was an ecommerce site, they'd be able to afford a backup machine, along with other niceties that come along with making a box mission-critical.
The crashes a few weeks ago were more or less because of administration mistakes.
You like cherries?
How you like them cherries?!
I run literally dozens of e-commerce site off solaris 2.5 and 2.6 using e450's up to an e10K
for the portal, the stuff is solid and dependable
I submit that configuration is ALL and can make or break a business, be it router/network config or local system config.
As an archive site dejanews has lots of database servers too, all of which also run linux. I think their server farm is up to ~=200 machines now. I belive that they easily quailfy as 1,000,000 hit a day site. Ironically, if you take a look at their homepage, IBM appears to be one of their main sponsors.
"Maintaining a FreeBSD system is a real pain in the ass"? I beg to differ.
I've run several FreeBSD systems in various roles from high traffic Samba, HTTP, and SMTP servers to the not so high traffic internal servers and even on a few desktops. I've _never_ had a problem with any of these systems.
A FreeBSD install gets me a fully funtioning machine and source for all of Userland and obviously the kernel. I use CVSup to sync my source with the latest patches/updates/etc... everynight. Upgrades are as simple as 'make buildworld; shutdown now; make installworld; exit'. I've the latest system built from latest sources (That's Userland and KernelLand) - You can't do that with Linux, at least not with any distro I've ever seen.
The ports collection makes installation of 3rd party software a breeze. The 'pkg' system on FreeBSD keeps track of dependencies, allows me to build from source, and always gives me a perfectly functioning binary in the end. The 'pkg' database is also _far less fragile_ than RPM.
I ran Linux for years before I made the switch to *BSD. I left Linux because administration was such as pain. Since I've run FreeBSD I've been completely free of most of my administration tasks, these boxes "Just Work".
Some months ago I was 'portunate' enough to be
part of a team managing a major e=commerce
venture running OpenMarkets Transact on solaris
2.6. Admittedly the machine was initially poorely
configured, but the thing was still seriously
unstable, and it was running on two sun E3000s'.
Frankly for mission critical stuff I would go for
linux solely because I know the kind of probs it
can have, and canget community help to solve them.
It has NEVER crashed due to the web traffic. It has crashed a few times (4-5) during the last two years due to old, cheap and flaky hardware, but that's hardly something you can blame the OS for... (the little downtime there is doesn't warrant me buying new hardware yet)
At the same time, the machine serve as a mailserver, and as my personal toy machine for misc. web projects that involve frequent gcc compiles etc.
So Linux can't handle millions of hits a day?
The problems Slashdot have had in the past, I suspect are more a result of extremely dynamical pages. Most sites stick with a bit more static content.
Note that the above is the average over a ~2 month period (ie peek values are much higher).
Of course, I wouldn't want to run the above server with Apache or similar. :-)
Apache? Not quite...
internal-roxen-roxen gives a hint...
I wish I had the chance to work on a million+ hit web site but so far i haven't been exposed to it. At the other end of the spectrum, I have setup a 386 with 8 MB of ram and a 90 MB drive ... it acts as a firewall for 50 PC clients and blows the doors of the Windows solution that was there before (both speed and reliablity). My stripped down Debian install uses only 33 MB of disk space! I would think that efficiency on low end hardware should translate into good performance on high-end equipment as well.
I think that the problems would be more due to the limitations of the hardware, rather than the limitations of Linux..
If you had a Fibre-channel Linux box with 8-way SMP, plus multiple Ethernet cards and all the other goodies, I'm sure Linux would do just fine. I understand that there are problems when you go even higher, like 16+ processors -- but it's being worked on, of course.
I know it's commercial, but it's built on the same technologies(for the most part).
I once administrated windows95.com on a P90 with 32MB of ram running Apache 0.8.X, using BSDI, and it ran like a champ topping our T1 out at around 2 million hits/day(peaked at 4M/day frequently).
There was quite a bit of kernel and apache tweaking involved, but it makes a statement for these technologies.
Gee, I thought IBM was on our side.. How things change overnight...
Nobody's going to bet their business on tuning a server to meet the needs of the clients? I can tell you one thing, if that business gets popular, then they're royally screwed.
You wrote:
"If you run a commercial UNIX, you can expect close to linear scaling with multiple processors as everything is done very fine grain."
I have to disagree with you; unless you are running code that fits almost completely into the L2 cache you will not get anywhere near linear scaling with ANY OS - I don't care which one (yes, huge L2's help a lot, but I'll bet Linux 2.2 would also fly on an 8-way Xeon with 2Mb L2 cache each).
You have to remember that with shared memory SMP the processors have to share access to the main memory; and while the memory bandwidth does not get reduced to 1/N (N=number of CPU's) it does approach 1/N more closely with every CPU added...
Depending on the locality of reference, ignoring kernel locks, a big (>= 512k) L2 cache will give you about 90%-95% cache hits for workstation loads, but for server code you spend more time shuffling data for I/O or doing database searches; hardly L2 friendly. I doubt that more than 70%-80% of memory accesses are served out of the L2 cache on a big server.
Processor contention for memory resources can cause a drastic slowdown; a friend of mine once saw a dual processor box perform at 60% (yes, slower!) the speed of a slower single processor box! (there is an article about that on my site)
--------- Webmaster, http://www.cpureview.com and
IBM's not even on their own side. They're so big that their right hand doesn't know what their left hand is doing, much less how to give it a .. er.. hand. If you need more proof of this than the contradictory statements coming out of their executives, I have one word for you:
OS/2.
Enough said?
No, that's neither a kernel nor a database problem. That's a filesystem problem, showing the limitations of the ext2fs filesystem. A database shouldn't have to implement its own filesystem just because the current one is extremely limited. On BeOS they don't have to either =)
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
FreeBSD != Linux.
They have TONS of differences. Just because FreeBSD can do something, does not mean that Linux can do it (and vice versa). FreeBSD generally seems to be better at being a server. (witness cdrom.com's record-setting FreeBSD server).
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
I'd think that'd be a clue.
Posted by OGL:
Have you considered your phone lines/isp? It's fine on this T3 connection.
-W,W.
Posted by FascDot Killed My Previous Use:
/. are:
I have to note that the two slowest things about
1) adfu.blockstackers.com is INCREDIBLY slow. I usually have to hit cancel to be able to see the page (which is fully loaded otherwise). This is unacceptable in a banner.
2) images.slashdot.org sometimes does the same thing.
AND, since I'm already on the topic of banners:
/. Feature Idea: Since I have icons turned off, I'd like be able to move the banner to right next to the page title (assuming I have to see it at all). This would give me an extra inch at the top of the window allowing an extra story.
Posted by Nr9:
the hotmail web page is run on free bsd while their mail servers are run on linux
Sun doesn't even make these any more. Are you sure you can
qualify this as a 'major' e-commerce venture?
Lurking at the bottom of the gravity well, getting old
F0 07 C7 C8
FreeBSD is no more reentrant in the kernel than Linux 2.2 is. About 96-97% last time I checked, which may as well be "not at all".
Security is OpenBSD's forte.
FreeBSD happened to be better able to handle brutal loads when Yahoo was setting up shop.
"Nice" SMP is bullshit. Either kernel threads work or they don't. That's all there is to it.
(well, as of 2.2; SMP on Linux 2.0 was poorer)
If you want ultimate scalability and reliability you buy a mainframe. If not, you're taking your chances with hardware that is "only" singly or doubly redundant.
Remember that what's inside of you doesn't matter because nobody can see it.
IBM ought to be in the business of pushing their mainframes for the real high end, where no one else has anywhere near their name recognition ("mommy, what's a Unisys?"..."isn't a Tandem a bike with two people on it?").
This guy probably has to sell AIX (poor SOB) and doesn't want to admit that its JFS is the only good thing about AIX. (yes Virginia, DFS sucks)
IBM ought to concentrate on
1) mainframes in back, and
2) Netfinity boxes running tuned Linux kernels in the middle;
and let the desktop maroons choose their own OS.
I.e. put Linux on a Thinkpad and sell it, now.
Fuck, I'd buy one. (the VArBook 120 is $4000 and my old PowerPC notebook has booted its last)
Stupid ass infighting. IBM is so close to (yet so far away from) the unity it needs to dominate without manipulating the market (ala Microsoft).
Remember that what's inside of you doesn't matter because nobody can see it.
No, but you can give a "thank you Sybase" if you want to run a commercial RDBMS for free in deployment.
Unlike Oracle, Sybase offers technically adept users the chance to run an older version for free, for full deployment (unsupported of course).
Personally I don't see anything wrong with running the site on MySQL. Oracle is not bug free, Sybase is not bug free, DB/2 is not bug free, and Informix is not bug free. They're just big and baroque, and have been tested for longer.
Remember that what's inside of you doesn't matter because nobody can see it.
First of all, it's actually Free. Period.
Second, it handles transactions.
Third, it is actively developed.
Unfortunately, it has some shortcomings and is not "industrial-strength" in some areas. Plus it's wicked slow. (wait, a wicked slow RDBMS -- that's redundant!) However, for applications that are multiple-read, multiple-write, it is the only choice that's Free. MySQL is appropriate for multi-read, single-write, but not transactions at the enterprise (eg. bank, hospital) level.
Remember that what's inside of you doesn't matter because nobody can see it.
A quick run of limit shows, and sets, per user, the maximum number of file handles one can have open.
/proc/sys/fs
root@dogbert [~] cd
root@dogbert [fs] cat inode-max
8196
root@dogbert [fs] echo 10240 > inode-max
root@dogbert [fs] cat inode-max
10240
root@dogbert [fs]
Is that what you consider a "recompile?" Perhaps you're berating older versions of the kernel. Would you serve a million hits off SunOS 3? Is it relevant?
Linux and current hardware of the Intel IA32 or Alpha or Sparc generations have no problems saturating a 10 Mb connection for an entire day, week, or year. Of course, as you add in server side dynamic content, memory bandwidth and CPU requirements jump. It's the server manager's responsibility to make sure the hardware can meet the demands; if you need more CPU because you wrote scripts that require it, either tune your scripts or buy a faster CPU. Cluster your machines, do some DNS tricks, and farm out your web jobs. Perhaps invest in an SMP Alpha machine with plenty of I/O bandwidth an RAM. These are all server manager duties, and because so far the operating system has posed no problems, there is no call to "blame" the operating system for any of it.
I would love to debate your argument on purely technical terms, instead of hypothetical duties and theoretical loads, but you don't seem to present any technical arguments against Linux. In fact, I would classify all your current arguments as FUD, plain Fear, Uncertainty, and Doubt.
There's your standard FUD. "Oh no, there was a HUMAN ERROR in the past, what would happen if we extrapolated that to some fantasy future scenario and blamed it on the COMPUTER? Things would be mighty scary, indeed! Run from Linux, it is all things evil!" I'll tell you what would happen if Linux was serving an e-commerce site. Since this company has a brain or two, they have a backup web server, mirrored RAID systems, and do nightly backups. The main box goes down, the system administrators' beepers go off, they run into work, and see that the backup has automatically taken over web services for IP www.xxx.yyy.zzz. They get the original web server back on its feet, do some internal testing, sync the disk contents, and turn it back on. They go back home.Say the company wasn't so smart, didn't have secondary machines, and had to deal with not meeting performance. The system administrators, knowing the server scripts and hardware, would then look for things to optimize. Perhaps the machine needs more RAM, so it gets it, and they go home.
If the machine isn't running in an hour, and the company loses a million dollars, then the system administrators are told they are now free to find other jobs.
Tell me where, in those situations or others you may know, where Linux was a limiting factor here (I can only touch these hypotheticals because I have never seen Linux fail to serve content becuase it was suffering from any operating system problems). Mr. Sergeant don't include a single example of where a server didn't meet its expected load; I can only assume he's being compensated for his words by some third party. I would love to find out where he works, and what he's done with his life, but last I checked, his posted web address simply fetched a nicely printed Microsoft ODBC database connection error.
I almost forgot to include your "inodes and shit" argument. Since no one here knows what you meant by that, and don't appear ready to explain, I can only guess that you are talking about maximum volume sizes of ext2 partitions on 32-bit machines. These are 32-bit architecture limitations, and here's a tip: get better hardware; buy an Alpha. Perhaps you want smaller inode _block_ sizes? Perhaps larger? Use a different filesystem, or at least give me a single, detailed, technical argument in favor of your sanity.
"for the European market, Linux is the No. 1 operating system used for Web sites" Isn't this true for every market? "For the RS/6000 AIX market, 0% of the servers run linux." I can understand IBM trying to sell AIX but sooner or later they'll realize the pointlessness of it.
-- Too lazy to get a lower UID.
Only on Sun boxes, but it is true.
Plus, we have the (hands down) fastest network layer and process creation times.
--
Ben Kosse
Remember Ed Curry!
the mail servers are Solaris
--
Ben Kosse
Remember Ed Curry!
Linux handles 64-bit with the sole exception of physical RAM.
--
Ben Kosse
Remember Ed Curry!
...do not constitute a Linux flaw on our part.
--
Ben Kosse
Remember Ed Curry!
Now, let's count the total number of servers to power all of microsoft.com. Over 200.
--
Ben Kosse
Remember Ed Curry!
Really I've never heard of an NT server staying up during even 1000 hits an hour let alone an NT server running compilers and remote X clients during peak hours. When the corporate servers on my lan are crashing and burning the non commercial home Linux box keeps going. As long as VM doesn't run out.
IBM is way too big to focus 100% on something. OS/2 users know this first hand. The people within IBM who talk to the press are seldom technical people, they are seldom technically competant... This was one guy and linux probably competes with what he sells, he struggled and struggled to learn his product or get funding for it or something and now a free system is going to step on his turf? Not in his mind yet. This will get worse, IBM makes a lot of products that compete with linux and the people in those organizations are going to resist it. IBM also hires a huge number of "programmers" who only know Visual Studio and MFC, wait until you hear what they are going to say about linux when they start talking, it will boil your blood. This all just means that people are noticing linux and people are being expected to support it, it's really good news.
In some respects his statments are correct, I don't think most thinkpad customers would be into linux at this point. Even with gnome and kde and staroffice and all the great stuff, it's still a little nerd oriented. On the server end, linux is awesome but it still hasn't proved itself under heavy fire, IBM ran the olympic site with absolutely huge amounts of traffic and it all kind of worked. Linux has run some big sites but not that big, yet. The people who pulled off the really big stuff will be the last to accept linux.
Everyone here is taking the approach that IBM is some sort of monolithic entity, as if one person at IBM being interviewed is speaking for IBM as a whole.
I really doubt this is the case. IBM is, IMHO, more like a bunch of companies flying in loose formation. And fighting much of the time.
If IBM had any unity at all, OS/2 might have beaten Windows on the desktop.
This is just a way for one Pointy-Haired Manager to say something that serves two purposes:
I doubt this person cares at all for how IBM does; he'll trade that away just to see his division out in front.
His idea of "burying the competition" means taking market share (or simply destroying the market share) of a division down the hall.
This is why everyone uses Windows instead of OS/2, and why, if IBM as a whole doesn't do something to deal with its unity problems, it'll wind up like Digital is now, IF they're lucky.
Phil Fraering "Humans. Go Fig." - Rita
(currently testing something about signatures here)
I think there are some scalability limitations with linux right now. The IBM dude was referring to (I hope for his credibility's sake anyway) milllionS of hits per day..like amazon.com type scalability. [I believe they use Alphas and Digital UNIX]..
Of course, one would wonder how many people require that kind of scale, but businesspeople are (rightly, so) anticipating insane scalability requirements over the next decade as the herd "gets on the Net".
While DejaNews does scale well, I think he's probably talking about more than database access here: we're talking commerce transactions and decision support systems (with complex querys).
There are some things that are so large that you'd probably *want* to buy a Sun or Alpha box. Solaris is an excellently designed OS (and the source code is available to prove it)... Linux on those boxes probably would do a decent job too.
Whether this will be true down the road (Linux 2.4/3.0) is another story.
This also isn't just about scale, it's about reliability. Nothing is completely reliable of course, even amazon.com goes down the odd time.. BUT.. Slashdot *HAS* had some noticable reliability problems in past (mostly due to MySQL if I recall correctly). Sun boxes are known for their reliability vs. Intel boxes where unless it's a 1st tier vendor you don't really know the quality.
just my 2c
-Stu
Are you sure that those would be anywhere near 1million hits/day? I mean, /. is "only" 600khd, and I can't imaging that they would be hit as much as /. is.
Sure, they get a lot of ftp traffic too, but ftp traffic doesn't count toward hits/day. (Besides, ftp traffic is "easy" compared to most web traffic.)
SPF support for most open source mail servers can be found at libspf2.
There is more than a grain of truth to this. Linux doesn't scale to a 64-processor, multi-GB RAM, multi-TB disk systems. It is just that these same people don't realize how much you can do with a 4-processor, 1GB RAM, 200GB disk Linux box, and how many PC users would think that is a large server.
SPF support for most open source mail servers can be found at libspf2.
Slashdot had serious kernel level problems when it's hits started increasing. AFAIK CT had to do a recompile to fix it. Noone is going to bet a serious business on that.
/.) but I wouldn't bet a million hits/day (average) web business on it. Imagine if Slashdot were an e-commerce site and had lost sales when it had it's "Problem week" a few weeks back.
Sorry, I love Linux (and
Matt. Want XML + Apache + Stylesheets? Get AxKit.
I helped Rob fix the inode problem (MAX_INODE in files.h) - the problem was that on hard hit sites MySQL hit the inode limit. (at least I think it's was inodes - that's why I said "and shit" - it might have been max-open-files).
It required a kernel recompile and a reboot. So there.
And to the other guy talking about static HTML - I'd trust Win95 or an Amiga to a static HTML site. That's not something to go by - any machine can serve 2 million static HTML pages a day.
Matt. Want XML + Apache + Stylesheets? Get AxKit.
IBM is positioning Linux as an NT alternative. It's only selling it in markets where NT is a serious contender. This is the low end of the server scale.
:)
IBM has 3 other high end server OSs. AIX, OS/400 & OS/390. I they put Linux in the same market they just fragment themselves more and loose money. Let them call Linux a low end server. It doesn't matter because people who deploy high end servers can dig through hype OK. If they couldn't NT would have supplanted AIX long ago.
Linux doesn't need to be marketed on the high end. It just needs to be sold and supported. VA Research ships 8 CPU boxes next month. Linux care supports them immediately and IBM gets to discover high end Linux when it owns the market
--= Isn't it surprising how badly I spell ?
www.realnetworks.com is runninga 12 on Linux
s t=www.realnetworks.com :)
Thisisarealoperatingsystemfromthefreeworld1.2alph
Linux users include Slashdot, register.com, and Deja News.
====
http://www.netcraft.com/cgi-bin/Survey/whats?ho
====
It's probebly Appache. but boy did they customize
--= Isn't it surprising how badly I spell ?
I don't think IBM is right, but I do understand how they came up with the statement. Managers understand surveys and rankings, its what they are trained to follow in the forms of trends and polls etc. What is needed to fight this FUD are concrete numbers from sources that management would find reliable.
For instance Dataquest among others routinely publishes rankings of market penetration of various computer manufacturers products in a variety of different markets, or of operating systems into various broad application areas. From this a manager can read these published results and point at a source which states that the penatration of Windows NT into the million+ hit per day web server market is 99 and 44/100 percent or whatever the results say. These results are made from a poll of approximately 10000 companies.
Here's where Linux runs into problems. Linux in general probably isn't even mentioned in the survey questionaires handed out to the various I/S managers who report these things to Dataquest. The sites most likely to be surveyed are also least likely to be polled by the Gartner Group, the people who run Dataquest. Slashdot for instance, while providing a valuable service to a rather large segment of the internet community doesn't register as a blip on their radar. It's just a few guys running a server as a hobby or if they're feeling kind a small business.
Even if linux results were tallied there is a good chance that it would be further sub divided by distribution. So for instance, Red Hat would be a seperate operating system, Debian would be, Slackware would be etc. This might be fixable if these companies referred to themselves as value added distributors of Linux loudly enough that Linux is the product and RedHat, Debian et al become distributors.
To help Linux be visible in polls it would be useful if a non profit company or co-operative were created which the various small businesses who rely on Linux could operate under. All of the information on servers, database size, web hits etc. would then be collated by this pseudo-company and delivered to the various fact finding organizations. Maybe small fees to be member (and to keep the company legally on the up and up) from which the entity can become an official client of one or more of the services.
I thought that DejaNews ran Linux.
They must take millions of hits per day.
Maybe the exec just doesnt know which sites run Linux because they dont get shouted about.
Deleted
I prefer my Blue Axe of Learning[tm]
Transactions are a different beastie. And I can assure you that there are *VERY* few million transaction sites out there. Is this database transactions? If so, I suspect /. and others could and DO scale to this level nicely already. If you're talking financial transactions, then there are really NO sites like that on the Web. Why do I claim this? At a US dollar per transaction, a million transactions are a million dollars. Now, each of these transactions that happen, just so happen to be a lot more than that. Does anyone honestly think that any of these sites, like Amazon, rake in much more than $500-750k per day? I sure don't. Don't base opinions of how an e-commerce site will fare with what goes on here on /. First of all, changes would be thoroughly tested out on a staging server before even being deployed on an e-commerce site. Second, they'd be fielding a hell of a lot more hardware than Rob's tossing at it. Third, they would have more admin staff than /. has. Fourth, they'd be using someone like InterNAP to guarantee bandwidth- much of the slowdowns we see when the site's under load isn't in Rob's immediate pipe, it's the fact that he's on only ONE backbone (Amazon and many other high-volume e-commerce sites use InterNAP because they don't use their own backbone, they use the big-8's and route accordingly to avoid MAE-East and MAE-West 90-95% of the time.).
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
"p.s. how many of you -can- make a flaky Linux installation if you want to? I can -- compile the kernel with -O6, agressive strength-reduction etc."
A trained monkey could do the same thing as you just described at any corporation you could name. Nobody in their right mind would do that- and that doesn't constitute "fickle" or capricious. To attribute what you just claimed to that is intellectual laziness on the part of the claimee.
However, having said that, I will agree to some extent that this is all due to a percieved fickleness to the OS. Something we're going to have to work on for IBM.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
If you deploy it in an application, it costs on a per user or per transaction basis. It would be prohibitively expensive for him to use it unless Larry (or one of his underlings) decided it would be good PR and an even better test of Oracle on Linux to have Rob use a "free" license.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
By the very nature of IBM things are pulling in different directions. I'm sure those IBMers who released Jikes and several of my co-workers at IBM knows better. Even some 'suits' know better. But a vast majority don't and are trying to protect AIX. As someone else posted IBM is trying to position, even if it's not really true, as Linux as mid-range OS and AIX as high-range OS (Unixwise at least .. way too much NT use here, but that's another story). It's marketing, folks...
>OS/2 1.x was an example of too much "unity" over at IBM --
Is that the last time you looked at OS/2? Wow, where have you been hiding?
>it ran well only on IBM MCA hardware,
You didn't look very deeply, did you!
>and included green screen emulators and other things that non-IBM shops did not want.
Excuse me, but "green-screens" are what run every large business that you deal with. Do you not own a bank account?
>I'm not sure if IBM really "unity", because they are just too damn big to sell an integrated solution like Microsoft does.
Too BIG to sell an integrated solution? An oxymoron! Try scaling things up a bit. Your desktop hardware, software, network card, cable, hub, local server, wide area network, enterprise servers, global network with sub-second response time. Think man! How does IBM manage the employment of over 220,000 people worldwide; on NT servers? HAHAHA I laugh at your smallness!
>If they did, it would be centered around the big iron mainframes, because that's where their money is. (And that is what they tried to do with OS/2 1.x.)
Sorry. Not very much money in mainframes today. Services is where it's at. Installing and manageing all those desktops, servers, and networks that "you people" use. Something that Microsoft is just too SMALL to understand.
>Just the fact that they sell a Unix solution and are considering x86/PPC Linux prove that a little disunity is good over at IBM.
Yes, a little "disunity" is good over at IBM. Which is what we're seeing here.
>(Although, let's face it, most of their revenue is from the installed base.)
Wrong, wrong, and wrong. You haven't done your homework!
Ken
Take this from an IBM'er who previously has worked on the "outside" for MOST of his career;/ lw-03-penguin.html
There is no "They" from IBM. You won't get a consistent opinion from everyone at IBM, merely different opinions from different heads of divisions. The clown who said this will be corrected, eventually. Assume the opinion comes from a much smaller company who has never dealt with the Americas. If you want to know how HUGE companies appear to the public, read this;
http://www.linuxworld.com/linuxworld/lw-1999-03
Ken
Ahem;
Microchannel was the pre-cursor to today's video board design: Peripheral device intelligence reduces the load on the CPU. IDE vs. SCSI, no? It was a good idea.
The only bad idea was that they tried to make money by licensing the design. VHS vs. Beta?
Ken
actually, "maintaining" a FreeBSD system is a piece of cake. /usr/ports is a really nice way to install third party software. cvsup is a simple way to keep up to date with the latest source revision, and make buildworld/installworld is a piece of cake. If you think that FreeBSD is hard to maintain, then I'm guessing you refuse to read documentation. The current FreeBSD distribution system rules.
---
---
we stand in life at midnight, we are always on the threshold of a new dawn.
This is interesting could anyone expand on this??
While not an expert here is my thought...
I have heard that high end databases use there own filesystem for the database, and therefore don't have problems with clusters or inodes. On the other hand I think MySQL uses the linux filesystem for its database, and therefore is subject to the limits of the files system. So, rather than being a kernel problem this would actually be an issue of using a database not designed for the load being put on it.
Taken straight from the article: "We see Linux as a server phenomena right now more than as a desktop phenomena," and then several paragraphs later - "No one runs large, million-hits-per-day Web sites on Linux" So he thinks it's a server phenomena but can't handle a million hits per day? This doesn't make sense to me. It's like saying "IBM is an awesome company...nobody uses any IBM equipment or software for anything spectacular."
I think there are some scalability limitations with linux right now.
I've got a few Linux boxes here getting 3M hits per day EACH!!! No problems...
To me, /. is perfectly fast. Sure, there have been a few reliability problems in recent weeks, *but* remember that /. was, up until very recently running on a dual PII/266 w/256MB RAM. And remember that /. is on a T3, (considering the number of cable modem and ADSL hits /. must get makes a T3 look pretty slow). All things considered, /. is not an eCommerce server. If it was, profits would mean a way faster server (probably AXP w/Linux), a backup system and more people to write scripts (and to debug them, etc.) Although Linux was origionally made to a be a PC UNIX, thus the lower end of UNIX, it's scaled quite nicely. I'd bet a multi-million a-day website running Linux, if I had like a quad Alpha 21264 with a good backup server. Although there are some scalability problems, Linux scales WAY more than SCO UnixWare or Windows NT (it's immeidate competitors, along with possibly Solaris/Intel). But I'd be happy if IBM would just include Linux as an option with all NT-class machines that it offers, as Linux is proven for these machines and is much more scaleable than NT is. Tim
That's 5,238 transactions a day.
You need to remember something though.
/. != Linux.
Yes Slashdot is running on Linux. But CT is always playing around with it. Also, as the kernels improve, so too should performance.
Slashdot tested in on it's new box fairly solidly, and the errors received were more due to bugs in the code for the pages than from any load-up on the kernel.
I propose that we attempt to Slashdot Slashdot.org with all our might for a solid week. Just to test the stability of the box under heavy-duty loads.
The only reasons the IBM people are saying "nobody runs it on million-hits-per-day servers" is because:
{Pokes CT in the shoulder}, HEY ROB! YOU LISTENING?
Chas - The one, the only.
THANK GOD!!!
Chas - The one, the only.
THANK GOD!!!
It sounds pretty cool - I wanna job there :-)
support gun control: take guns from cops
If you use a proxy you wont ever have to see adfu again (I don't). Rob isn't going to advertise this - he makes money off the ads, after all. You're expected to know better. For those that don't, well, deal with the ads.
I haven't seen a adfu ad for months; /. is really fast for me, all the time. Well, during peak hours it might be a little slow.
support gun control: take guns from cops
I know Oracle sure as heck is NOT free for linux, unless you just want to download it and play around. That's about it. Has anyone used postgres for anything realy serious? It seems to be under very active development (they seem to have regular releases) and the developers sure do seem fired up about it :-) I haven't done much with it other than make a few tables ... is it really a suitable replacement for high end DB's? or maybe "middle end", where oracle would just be too much?
support gun control: take guns from cops
[Names changed to protect the innocent.]
$ ll access_log.1 access_log.2
-rw-r--r-- 1 root root 215507840 Mar 20 01:05 access_log.1
-rw-r--r-- 1 root root 214266973 Mar 19 01:06 access_log.2
$ wc -l access_log.1 ; wc -l access_log.2
1083612 access_log.1
1079518 access_log.2
$ uname -a
Linux server 2.0.36 #1 Tue Nov 24 05:29:56 PST 1998 i686
And so, please pay no attention to the funny man with the "Thimk" license plate -- it is mearly more silliness from the people who brought you Microchannel(tm).
-}Creon
I dunno. I doubt a Pentium/90 sitting on a 128K ISDN line could serve millions of hits a day. But slap Apache in round-robin proxy mode in front of a cluster of Linux machines and you probably could.
Right tool for the right job -- don't expect a smallass Pentium to go up against big iron, whatever the operating system. But provided you're willing to throw the right hardware at it, Linux can compete against the "big boys" just fine.
Nothing worth doing is worth doing today.
So, with "inodes and shit" you meant, that the standard kernel configuration as shipped by some distribution imposed a (let's say arbitrary) limit that could be changed easily and didn't even require down loading patches or actually fixing bugs ????
...)
Would this also mean that if Linus (or whoever worked on that piece of code) would have set a higher limit then you would consider linux suitable for a million pages a day ?
Is it just me or do your statements just merely indicate that Rob and you spent a night hacking and cursing until you found why MySQL bombed and that you are pissed that it was just because of a silly, unnecessary limit set in the kernel?
I'm glad you helped Rob getting Slashdot to work but please let your unpleasant experience not obscure the view on the whole issue. You can say you wouldn't run a high end web server on a plain vanilla RedHat/Debian/Suse/@&^%$&^% install, and everybody would agree. You could argue that there should be an easy to understand kernel configuration option "Configure for server [y/N]" that would raise some limits, and most of us would agree.
The bottom line I seem to be getting here is that a lot of people got linux based high traffic web servers working after some tweaking, but there appear to be different opinions on how much tweaking is acceptable. This could be easily solved by releasing a distribution that is configured for high end servers (higher limits, some desktop stuff removed, basic server packages preconfigured and running, serial console, firewall,
Her serveer is another Linux box which handles a huge (ludicrous even) number of hits.....
Just to get some more names, a french portal called voila runs Linux (look at the icon at the bottom of the page near the IExplore one). And so does DejaNews.
But don't we just forget FreeBSD which has what might be the fastest tcp/ip stack for that kind of hardware. Just look at this nice bast*rd of ftp.cdrom.com who didn't let me in just because there were already 3600 users logged in. I guess this beast just explodes regularly most records and NT clusters Microsoft (r) can find.
--
Memory fault -- brain fried
What do you wanna run ? MacOSX ?
muuuuaaaaaaarrrhh.
sorry, just kidding.
--
Memory fault -- brain fried
www.xoom.com
they get a million hits a day easy and they run
redhat linux
YOU CAN'T FIX STOOOPID
My main gripe with slashdot's performance is the rendering speed. Massive tables just are dog slow. Not that I'm all that much better. It does go to show that performance isn't just about cpu cycles. You can make a linux very slow, and you can make NT scream (shudder), it's just easier to get good performance when you're working on something that isn't based on VMS. O'Reilly has a decient book on web performance tuning if you're intersting.
The power of technology is manifest in how it is applied within the social matrix.
I do this every day. I run a site with millions of hits a day. It runs on Linux. It doesn't fall down -- day after day after day. High hit loads all the time. When we see 13 hits per second, we think that's low load. :^)
This
My company runs very large scale web sites on Linux right NOW. We
get millions of hits a day, and Linux handles the load perfect, without
complaint, and without problems or crashes. In fact, we use an Irix box
for a file server on some of the sites and THAT machine crashes
frequently.
As far as we are concerned, Linux is the best investment our
company has made. The machines it runs on are cheap and plentiful. The
operating system is low cost and very fast and reliable.
I think IBM is wrong when they say it shouldn't be used in these
situations.
In addition, we also use Linux on the desktop for almost all
employees. (There are a number of people in the company who need Word and
so they don't use Linux.)
This
Hitachi? Don't they make gas grills?
I've finally had it: until slashdot gets article moderation, I am not coming back.
They want to maintain the image that Linux is for "low end to mid-range" servers. That way they create a perceived demand for their proprietary OS in the "high end" server market.
Slashdot had serious kernel level problems when it's hits started increasing. AFAIK CT had to do a recompile to fix it. Noone is going to bet a serious business on that.
Nobody is going to bet a serious business on an OS where they can tune the kernel to meet their particular needs that easily?
You don't have the slightest clue what you're talking about.
A serious business won't rely on an operating system where you *CAN'T* tune the kernel that way.
I work for FedEx. We have recompiled kernels to meet specific needs thousands of times.
Are you meaning to suggest that FedEx isn't a serious business? I bet we do more business every day than your company does in a year.
Scalability? Linux is more scalable than most OS's. my friends. We have Beowulf for PVM fun..
and we have SMP, for 4-8+ way fun.
Oooo, 8 whole processors!
It gets awful after 4, sorry; circuit-switched SMP is like that, and always will be.
Solaris scales well up to 64 processors right now, and the stuff in development is orders of magnitude beyond that. (Can't say more, under NDA, sorry.) HP-UX is the same way.
Linux kicks ass on the low-end box, and when you put lots of low-end boxes together in a Beowulf cluster you can kick the pants off the big boys in terms of bang for the buck, but let's not do ourselves the disservice of calling it all things to all people on all platforms. That's Microsoft's style.
"does 64 processors" and "does 64 processors noticably faster than 8 processors" are two different statements.
Want to disagree with what I say on it? Fine.
Here's what Linus Torvalds has to say on it, from his section in Open Sources:
Symmetric Multi-Processing (SMP) is one area that will be developed. The 2.2 Linux kernel will handle four processors pretty well, and we'll develop it up to eight or sixteen processors. The support for more than four processors is already there, but not really. If you have more than four processors now, it's like throwing money at a dead horse. So that will certainly be improved.
But, if people want sixty-four processors they'll have to use a special version of the kernel, because to put that support in the regular kernel would cause performance decreases for the normal users.
In other words, Linus says EXACTLY the same thing I do! So grow up and stop doing Linux the disservice of claiming it's all things to all people. That's the tactic that's shooting Microsoft in the foot right now.
Well, there you go. Please do us Linux lovers a favor and forward this statement to the press and make that schmuck eat his words. I wouldn't know. I've never worked on IBM's unix, only Solaris, HPUX, SGI's, DEC's, and Linux, but this was obviously an intentional media slam on Linux in favor of IBM's unix.
-El C.
Um... DejaNews?
OS/2 1.x was an example of too much "unity" over at IBM -- it ran well only on IBM MCA hardware, and included green screen emulators and other things that non-IBM shops did not want.
I'm not sure if IBM really "unity", because they are just too damn big to sell an integrated solution like Microsoft does. If they did, it would be centered around the big iron mainframes, because that's where their money is. (And that is what they tried to do with OS/2 1.x.)
Just the fact that they sell a Unix solution and are considering x86/PPC Linux prove that a little disunity is good over at IBM. (Although, let's face it, most of their revenue is from the installed base.)
--
Business. Numbers. Money. People. Computer World.
FreeBSD doesn't NEED a distribution. The whole point is that BSD is scalable, meaning that you can run it on ANY kind of hardware without gratuitous apps eating up space. Coming from a BSD background, I can't understand what all you linux folks are going on about with distributions. What difference does it make? Why do you want preloaded software?
Anyhoo - administration is simple for anyone with a moderate unix background. And, FreeBSD has been proven again and again to be able to handle higher loads with greater stability than Linux.
17 linux servers running DejaNews? What does that prove? 17!??
-lx
The (lack of) speed of this site (if any) is due to bandwidth problems and pages being big and requiring the whole page to download before most (all?) browsers display them (because everything goes inside a big table). It has nothing to do with the site's operating system.
It is very sad to see that many suits seem to think the way you do and take decitions based on irrelevant stuff.
AFC.
And let us not forget: the godhead search-engine Google runs on Linux. I bet they get more than a lousy million hits a day.
Slashdot does 600,000 page views per day. HIts and page views are completely different stories (depending on how you count hits... and how you count pages, but I won't go there). The site I work for does 1.5 to 2 million hits per day (on Linux), but our page view count is much lower.
e;
No, most major databases either require or have the option of using raw devices so that they can perform to the high demands of the system. Why put another layer between the database and disk if all of your multiple disks in your raid setup are used by the database? It's ignorant to do so.
It has already been mentioned that high-end production systems always should be customized. The max inode and/or max open files is just one example of such a configuration. Ever heard about configuring shared memory limits on a Solaris box for a database? I think changing those requires a reboot also. If so, what's you point?
Yet.
I think, IBM is afraid of LINUX as a potential competitor for AIX.
I agree with the posters who have already said that IBM probably doesn't give a damn about Linux competing with AIX. I believe that the reason they're offering Linux as an option is because of the unbelieveably huge amount of publicity and hype that has recently been thrown around the OS.
(I personally believe that most of the publicity and hype is well-deserved, but that is of course just an opinion)
--
Paranoid
Paranoid
Bwaahahahahaa.
This statement shocks me. If they ran linux (which nmap confirms), why would they have such a crappy linux port of realplayer?
Those people piss me off.
--
Paranoid
Paranoid
Bwaahahahahaa.
Yeah, and I have a wireless satellite-dish feed, which is technically a cable modem (if anything it adds a step or two) with 56k outgoing bandwidth. This adds a couple more steps and probably a whole new dimension of interference, yet the only problem I have is the adfu banner not loading, and thats no problem as I hate ads anyway. (and is NS 4.x specific?)
And within the last week, I've read slashdot at each of the 24 hours a day contains.
Seriously, check your own setup.
--
Paranoid
Paranoid
Bwaahahahahaa.
/. takes about a minute to load with Netscape 4.5 and it takes only a few seconds with Mozilla
The people at Realnetworks run linux
If you want an Alpha, buy an Alpha and put Linux on it. :-)
About slashdot's stability, read rob's explanations of what happened. The instability usually falls under one of two categories:
1. mySQL crashed from some bug in it.
2. Rob saying, "noone should ever give me the root password to anything"
Solution to 1: use a better database. mySQL isn't built for great scalability or much complexity, though the way that it is handling slashdot on not very much hardware is a decent testiment to it's scalability.
Soltuion to 2: Don't give rob your root password. I.e., make sure to hire qualified administrators. This isn't a dig against rob, but I think that even he'd say that it isn't a fully qualified administrator for this kind of site, though has learned a great deal and is learning. (I'm not calling myself a qualified admin either, btw, so this isn't any sort of attack.)
They laughed at Einstein. They laughed at the Wright Brothers. But they also laughed at Bozo the Clown. -- C. Sagan
I dunno, I run 4 FreeBSD boxes and think its the greatest thing since sliced bread but Linux drives me insane and mystifies me at times.
Its all in what your used to I think, not really an issue of true useability but rather human limitations in how we 'stick with what we know'
I don't think /. is the slowest site I visit. It's not *terribly* fast, but a lot of times it's faster than my.yahoo.
The argument about whether or not Linux can serve a million hits a day sort of misses the point: a million hits a day, in the grand scheme of things, doesn't require very high-end hardware. I'm pretty sure a two-processor Sun UE 450 could handle that with no problem, and that's a *workgroup* server. You'd need throughput of about 4MB/second for the peak, including processing....yeah. Should be doable.
-k. ^-^ ^D
I doubt it... Yahoo runs all its web servers on FreeBSD, so I'm certain someone's using Linux in the same type of environment.
yup, this site is lightning fast from home
and shit slow from work, so chances are i would say you have a shit isp... get a new one.
so what if its a cable modem, they can suck to.
"Computers will never truly be free until the last windows user is strangled with the entrails of the last mac user."
Just takes a bit of learning. I learned OpenBSD perfectly from Linux in about two days (with some help from EFnet#OpenBSD).
If its a pain in the ass maybe you should concidering reading some of the documentation.
"Computers will never truly be free until the last windows user is strangled with the entrails of the last mac user."
Bah, BSD's arnt really any harder then Slackware.
A bit differnt, so you might get a little confused for a wee bit, but with a little help from docs and mabye some internet places you get the hang of it real quick.
Try OpenBSD, its got the best security in the business. Thats what i use =P
"Computers will never truly be free until the last windows user is strangled with the entrails of the last mac user."
Boy, some of us arnt afraid of a prompt and a bit of source code.
Hell, since when is typing "Make" to install a goddam package so hard?
Don't be troll, sucka!
"Computers will never truly be free until the last windows user is strangled with the entrails of the last mac user."
I wont forget it. That's because my facility is hosting www.panamsat.com. The company who's sat blew it's zap. How is this relevant? Well the server that site was running on that day was a pentium 166 with 64mb of ram running Linux. This server is shared among other clients such as Ritz Carlton Hotels. On the day after the breakdown that 'puny' box handled 1.2million hits for www.panamsat.com alone. We routinely host sites on our shared servers which total 1 million+ a day. Were about to add yet another one that does 1.5m a day by itself.
I wish these folks would check their facts before making stupid statements.
An AC wrote:
> In the prime time (about 7 - 9 pm) slashdot slow
> to the point of being unusable.
Oh cool. Between 7 - 9 pm in what timezone? Better, on what continent or country?
If Slashdot is slow, there can be a lot of reasons
for it. It's not that fast from here but I'm on
the other side of the globe and that's a lot of
hops. Slashdot also has very dynamic content and
who knows what those scripts do. Tables are also
used a lot and current browsers are slow at
handling them. And then there is also the SQL
database. While they are easy to use, nothing
beats well made scripts that don't need standard
databases.
IBM has many parts, many of which are doing the same thing, few of which even know about the existence of these other groups. Furthermore, many of the IBM folks I've dealt with over the years have a severe case of NIH (Not Invented Here) syndrome. (If it's NIH it can't be good.)
I can easily imagine an AS/400 manager thinking that Linux is his chance to make his machine even more relevant to the world. I can also imagine anyone who sells CICS or IMS or other highly proprietary software seeing Linux as just another enemy. (Someone in this position would be an ally of Java and EJB, however.)
IBM will never be "on our side". Their interests may, in some situations, align with ours. Let's hope there are many such situations.
I strongly recommend "Big Blues, The Unmaking of IBM" by Paul Carroll. It shows how a combination of arrogance, incompetence, and infighting led to the dominance of Microsoft at IBMs expense. You can't read this and not begin to see Microsoft emulating IBM. The book is also fascinating because it shows how a big, scary company can do such a consistent and thorough job of shooting itself in its foot over an extended period of time. Due, primarily, to "not getting it".
In reply to the comment:
"Does anyone honestly think that any of these sites, like Amazon, rake in much more than $500-750k per day? I sure don't."
Take a look at Travelocity, it's done $11M/week more than once, do the math yourself. It's not a million transactions, but certainly larger than a few hundred thousand bucks a day.
On the other hand it's not Linux. The web part is Irix, the backend is IBM's TPF OS on S/390.
Umm.. lets clear some of the air, folks. :-)
;)
the ext2 filesystem is the fastest thing around. Thank you for playing. Put SparcLinux on a Sparc20, and Solaris on a Sparc20, and run web hit tests that are filesystem intensive. You'll see the HUGE difference between UFS and ext2. And then there's the fact that Network performance on Sparc/Ultra Linux blows the panties off of Solaris..
Reliability? Reliability and availability are two different things. If you do not plan for availability, then you have some down time. Period. Slashdot is a good example. However, given that they have NOT planned for availability, it is up more often than I would expect, given the heavy load. And is reasonably reliable.
Scalability? Linux is more scalable than most OS's. my friends. We have Beowulf for PVM fun.. and we have SMP, for 4-8+ way fun. And we have the good old Linux Virtual Server project, whose functionality has almost completely been folded into the new kernel. With that you could scale to hundreds of servers. No, big scalability is techinically feasible, and extremely plausible on Linux.
Need I go on? There's quite a bit more I could spout about, but I'm sure the drift is floating in the right direction.
I don't know what kind of krak these folks at IBM were smoking, but they are dead wrong about Linux. AIX is a dead fish in the water compared to Linux, but IBM is still flopping around with it in the noon heat. Remember, IBM as a 'whole' doesn't really like Linux.. just a few divisions of it which managed to get the corporate nod. These folks are likely from the opposing camp.
I say ignore 'em and march stolidly on towards technical excellence. Or something like that.
---
"Eternal vigilance is the price of Freedom."
"mommy, what's a Unisys?"..."isn't a Tandem a bike with two people on it?"
:-)
Yeah, and who the heck are these "Hitachi" guys, anyway?
Sometimes truth is stranger than fiction.
Here's a list of products from their web site's Product Finder:
Semiconductors & IC Home Page, Microcontrollers & Microprocessors, 8/16-bit H8 MCU, 32/64-bit SuperH MCU/MPU, Flash MCU, SuperH enabling Windows CE, Memory ICs, Flash Cards, LCD Drivers/Controllers, QAM Demodulator, Logic ICs, Optoelectronic Components, RF Power Amp Modules, Discrete Devices, Packaging, Developer Information, Mass Storage Products, CD-ROM, DVD-RAM, DVD-ROM, Hard Disk Drives, Multimedia Recorders, MPEGCam, M2 Multimedia Recorders, Handheld PCs, LCD Projectors, 8mm Camcorders, ATM Products, Accessories, Analytical Instruments, Belt Sanders, Brake Hose, Bread Makers, Broadcasting & Professional Cameras, Chemical & Nuclear Plant Equipment, Chiller Heaters, Circular Saws, Color TV, Compressors, Dealerboard Products, Desktop Computers, Digital Printer, Drills, Enterprise Professional Services, Fans, Full Size VHS Camcorders, Gears, Genetic Systems, Graphics Tablets, HMCA Diagnostic Ultrasound, HMCA Nuclear Products Division, Imaging Software, Industrial Video Products, Interconnection Materials, Inverter Welders, Inverters, Jig Saws, Large Motors, Mainframe Computers, Mainframe Storage, Mass Transit Systems, Massager, Microwave Plasma Etching Equipment, Mini-Notebook, Miter Saws, Monitors - Color, Multimedia Vision, Nailers, Notebook Computers, Open Systems Servers, Open Systems Software, Orbital Sanders, PBX Products, Planers, Power Plant Equipment, Power Semiconductors, Power Steering Hoses and Assemblies, Premise Wire, Printed Wiring Board Materials, Printed Wiring Boards/Assemblies, Printers, Printers - Color Laser, Programmable Logic Controllers, Projection TV, Projection Television Tubes, Reciprocating Saws, Rice Cookers/Food Steamers, Roll Mills, Routers, Semiconductor Materials, Shavers, Submersible Motors, Super TFT Color Displays, Super TFT Color LCD Monitors, Synchronous Optical Network (SONET), TFT Color Displays, Test & Measurement Equipment, TradeLink, VCR's, Vacuum Brake Hose, Vortex Blowers
Yeah, gas grills are prolley in there someplace.
[text processing courtesy of awk, btw.]
And no, I don't work for them.
Here's a list of products from their web site's Product Finder:
Yeah, gas grills are prolley in there someplace.
[(probably somewhat inacurrate) text processing courtesy of awk, btw.]
And no, I don't work for them. It's not my fault the list is so long. That was kind of the point of my last message...
I know XOOM uses Linux - definitely a million(s)-hit-per-day site.
Nitin
---------------------
The more idiot-proof you make it the smarter the idiots get.
I don't think that IBM really has any animosity towards Linux at all. In fact I think they see it as an opportunity to get a greater foothold in the 'low end' server market. The fact that a Linux server can handle a million hits a day is not relevant to IBM. They are trying to round out their product line while perserving their high end business. Its a good bet that their AIX based systems have a better profit margin. This provides a strong incentive for them to differentiate their servers in the minds of consumers. Are they going to say that it is better than AIX? No they are trying to leverage some marketshare away from NT based solutions offered by other major vendors. Linux is a good fit for IBM in this segment of the market, and that is how we will see it marketed.
on a slow day (friday) we got 1,667,218 hits. running linux (300mhz PII, load average 1.0)
:-) just the vast right wing conspiracy.
and its not even a linux site.. nor tech.
www.freerepublic.com
guess we're no one though
even more detailed solution to #1: Oracle on linux is free. Join the Oracle tech network and download Oracle8 for free. Can I get a thank you larry?
I thought I could!
I'm reading this in BeOS using NetPositive right now, and the pages are coming up way faster than NS4.x under Linux or Windows.
Plane tickets cost a hell of a lot more than books. Assuming an average of $300 per transaction, I believe (someone can check my work ;o) ) that works out to about half a million transactions per day.
Yes, this site appears to be very slow. It does take a while for it to show up in your browser. That happens for me both at work (T1 line) and at home (56K modem).
However, it has nothing to do with Linux, his ISP, or your ISP. CT seems to belong to the webmaster school that seems to be more graphic design oriented that technically oriented. Slashdot looks pretty good (arguable), but to achieve that look, it has tables embedded inside tables. It just takes forever to render.
I've actually written a cgi script once that crashed netscape by doing that.
I think slashdot needs to find a compromise between the look and the time to render. In other words, CT needs to tune its HTML - not his server or CGI (well, he may have to do that two, but that's another issue).
Isn't this the same people that demonstrated that Linux has supercomputer capabilities (they benchmarked a Linux cluster at the same speed as a Cray!). Now, I don't know about you, but it seems to me that they kind of self-destructed their argument.
My guess is that the person that made that statement (a subsidiary out of Germany?) just hasn't done his homework.
I run our sites on IBM machines running Linux without a hitch. Admittedly, we don't get millions of hits a day...yet. But we are getting close to that so we are building bigger machines...on IBM boxes running Linux.
`fortune -o`
Given the size of IBM, it's inevitable there will be a few clueless employees. Not only is he wrong (on purpose?), but he seems to be going against the "official" corporate philosophy.
You assembly coders make me laugh. When you want *real* speed, it's time to burn a custom chip. Slashdot could run on my pager!
I know a whole collection of people doing a good
million hits a day. These include big sites like
Dejanews.com. I wouldnt mind betting linux.org,
redhat.com, suse.com are getting those sorts of
figures off their boxes.
Anyway the big proof of the pudding this time is
amusing. Count the %age of porn sites running
Linux or FreeBSD. Compare it to those running AIX
Alan
Linux IMHO is a great desktop OS. It's the '95 of Unicies without the crashing. It has a lot of software and hardware support. Nice for some development, but then again, nice for development of Linux applications.
For a server, I would give FreeBSD (just like yahoo does) a try. Nice threading, nice SMP, nice performance. Easy security to setup (as most of it is pre-setup) and not ridden with scritps like the Linux distro's.
-spencer p
-
ping -f 255.255.255.255 # if only
What a idiot! Do a nslookup on www.microsoft and you will find out they are using SIX machines!
Doh!
is there a list of every time this happens? ;)
600k odd pages/day = about 7 per second
im sure linux could handle a lot better than that
Do your best, hope for the best, suspect the worst.
Why not? In exokernels, including ExOs, "redefining the filesystem" could be routine. A filesystem created for user use is not necessarily the best kind for database use (or web cache use, for another example).
Check out this URL for more:
http://www.pdos.lcs.mit.edu/exo/
I hope it's selective about those, we had to
throttle down the TCP and MTU params to stop
saturating the router with FDDI traffic. Normal
traffic could run fine on higher settings, but
when we started sending 5-6 10GB files at the same
time...
Gary
Dude, try this:s t=www.cybersites.com
http://www.netcraft.com/cgi-bin/Survey/whats?ho
I think that probably says it: It's running on Red Hat Linux and that's what his email address said too.
one of our web-chat servers regularly hits 1 to
1.2 million http requests per day, and has twice
reached over 2 milion. The machine is a single
pII-450 with 128M of ram, 40MB/sec adaptec scsi.
I don't think it's been down or rebooted more than
4 times in the past year. 50% of the data pushed
by it is dynamic, server-parsed html, or from
CGI's - IBM probably would rather see their "big"
machines running AIX doing the "big" jobs, so they
can aim the credit at their own products, instead
of a free product they just happen to support.