Booting A PIII System In .8 Seconds

gizmo_mathboy writes: "General Software has announced the fastest BIOS boot time on record. The embedded system was clocked at 0.8 seconds from system power-on to transfer of control to LILO. This was on an Intel SOYO motherboard (440BX chipset) running a PIII 400. I think the quote of the article is: 'This Embedded BIOS quick-boot operation allows the device to restart and resume operations well within three seconds -- the maximum amount of downtime allowed per year for a device that must support "seven nines" or 99.99999 percent uptime.'"

Comments

...but when booting Windows

[Alien Perspective]

It's only 0.6 seconds from power-on to BSOD!

I don't know about booting...

[The_Messenger]

...but I can certainly crash one that fast. AMD rul3z j00r 4$$!

yeah but

[-douggy]

Once i add some scsi , network, raid cards I can slow it down again :o)

Re:yeah but

[germinatoras]

No kidding, I'll bet that BIOS would spend at least 3 seconds waiting for my hard drive to spin up.

That's Great! The gaming industry will be pleased!

[JeremyYoung]

Anything that can be done to help those online gamers deal with crappy, untested game clients that crash their computers should help Electronic Art's, Sony's and other MMORPG game companies profit margins. I can see the recession turning around already, especially with gamers able to reboot so quickly.

that is fast but...

[mandria]

sometimes i would like to have a sec more to press the delete button to get into the bios and change a few settings.

Re:that is fast but...

[Roofus]

I would guess that you would hold down the BIOS setup button, and then power on the computer.

Re:that is fast but...

[saider]

This is designed for embedded devices, not your home computer. Therefore, you'd likely be hard pressed to find a delete button (or any other one for that matter) that would let you into the BIOS.

Re:that is fast but...

[Anonymous Coward]

So you don't get the "Keyboard not Found. Press to continue" error?

Just how am I supposed to press F1 without a keyboard?

Re:that is fast but...

[Amazing Quantum Man]

Stupid... I forgot about tags...

That should have been "Keyboard not Found. Press <F1> to continue".

Re:that is fast but...

[tswinzig]

sometimes i would like to have a sec more to press the delete button to get into the bios and change a few settings.

This is going to blow your mind, but you don't actually have to wait until you see the text to press DEL!

doink

Do we want this?

[Wind_Walker]

I mean, seriously, what's the big deal if it's 0.8 seconds from power to LILO? I, personally, would rather have a BIOS that takes a few seconds to check the RAM, auto-detect devices, and check SCSI drives before it tries to boot the system.

Why are the geeks so concerned about things being as fast as possible? What's an extra 5 seconds of your life? If it takes that long to get things done (and get them done right) then I'm all for a 5 second start up. Yes, it's killed productivity, but let's look at the math: 5 seconds per computer, figure about 1000 computers booted up per month (50 computers, booted up once per work day). That means you're wasting about (carry the 2...) 80 minutes per month waiting for computers to boot.

Now think about the time it takes you to get coffee in the morning; we're talking about 30 seconds or so (if you don't take it black) every day; that's 480 minutes of wasted productivity

Don't get me wrong, this is great, but is it worth the implementation? The numbers say no.

Re:Do we want this?

[Anonymous Coward]

What's an extra 5 seconds of your life

You just wasted 15 seconds of my life making me read your worthless post.

Re:Do we want this?

[Anonymous Coward]

>I mean, seriously, what's the big deal if it's >0.8 seconds from power to LILO? I, personally, >would rather have a BIOS that takes a few >seconds to check the RAM, auto-detect devices, >and check SCSI drives before it tries to boot >the system.

It says "embedded system". If you put a cpu in as a brake or steering controller into something that moves at any reasonable speed, you would like it to return service as soon as possible after a power glitch.

No, this does not mean that they make embedded controllers with crashes as a design goal. It means they want to make something that is as error-resistant as possible. Not for your desktop box, in other words.

Obtw: very few of those systems have anything like Linux or Windows on them, even though some people would like to tell you otherwise.

Re:Do we want this?

[ethereal]

This is an embedded system; it's probably being restarted by a watchdog or failed over automatically and thus you want it up as quickly as possible in order to minimize system downtime. Human productivity has nothing to do with it.

I imagine (not having read the article yet) that they've got some other layer to ensure that a machine with a failed device or bad RAM will not be booted at all, or perhaps with an active-standby configuration you could have the machine only run the extended POST if it is the inactive machine in the pair. There's no need to auto-detect devices every time; how often do you change devices compared to how often you boot? (OK, for Linux boxes with a UPS this is 1:1, but for those of us with unreliable power distributors or Windows boxes it's a little more often...)

Re:Do we want this?

[Telek]

Umm, anything like a brake and you do NOT WANT a power cycle. If that happens, you're toast. You go out of your way in any CRITICAL embedded system to ensure that you NEVER NEED TO CYCLE the power. So a 2 or 3 second boottime is negligable.

In any serious embedded device you want to make sure that you never never never need to reboot the device. So I think that the only cases when boot-time would be critical is things that need to be active within instants after a poweron. Things like power backup systems or ... I dunno. maybe bad example, but I can't think of much where the difference between a .8 and 3 second POST time is going to be a big deal at all.

Re:Do we want this?

[Anonymous Coward]

Mission-critical, but not safety-critical complex embedded systems, like process control or telephony.

Re:Do we want this?

[Anonymous Coward]

Embedded system...

Do you really want to wait 30 seconds to boot the embedded processors in your car, microwave oven, TV etc ???

Re:Do we want this?

[Junks Jerzey]

I mean, seriously, what's the big deal if it's 0.8 seconds from power to LILO? I, personally, would rather have a BIOS that takes a few seconds to check the RAM, auto-detect devices, and check SCSI drives before it tries to boot the system.

Now realistically, does a processor even need 0.8 seconds to run a memory test and scan a few devices? 0.8 seconds is an eternity on any processor made in the last ten years.

Re:Do we want this?

[Anonymous Coward]

Goto mp3car.com and ask those guys if they want a lilo prompt in less than a second.

Re:Do we want this?

[fishbowl]

>0.8 seconds is an eternity on any processor
>made in the last ten years.

On the *processor*, yes, but not on memory chips, address decoders, rom's, the things that are being checked and read during POST. Those devices are still slow.

To get a very good indication of how slow your memory is, try this memory diagnostic (x86 only):

http://www.teresaudio.com/memtest86/

Not only will this debug certain classes of problems that are otherwise hard to track down, but it will make you realize that the fast processor is not the whole story.

Re:Do we want this?

[John_Booty]

Don't get me wrong, this is great, but is it worth the implementation? The numbers say no.


I think one of the big turn-offs people have about comptuers is how long they take to boot up. Computers will be more popular with the "teeming masses" when it's something they can flick on like a TV.

And there's nothing wrong with that mentality either. Computers are tools. My machine takes like a minute to boot the OS... how many other tools or appliances take that long to boot? What if your TV took 60 seconds before it could perform its task? Or the phone, etc. Even cars don't have to be warmed up for that long in average conditions.

Of course, true geeks just leave their boxes running 24/7 anyway so boot times aren't terribly important. :)

Re:Do we want this?

[loosifer]

Yes, I want this.

The fact is, x86 systems are the only ones left that take this stinking long to go through POST (power-on self-test). Yeah, a $300,000 Sun box takes longer, but its POST is _significantly_ more complicated (try it in verbose mode sometime if you don't believe me). Check out how long it takes a Mac to start booting the OS--oh, that's right, it's barely even discernable!

On my machine, a dual-celeron with SCSI, on-board HPT controller, three SCSI devices, one IDE device, and 384MB of RAM, it takes longer to get through the BIOS than it does to boot my main OS, BeOS (of course, Linux is still slower than the BIOS).

I think that's silly. It's not like it can't be done faster, it's just that the x86 world depends on such stinking old technology that it isn't done faster. Everytime I see phrases like "extended RAM" and "640KB" I want to puke. Why haven't we dropped that like the dead weight it is?

So basically, yeah, I want this, and I want it five years ago. I know that this product is for embedded devices, but I want it for my desktop and laptop. Especially laptops--I hate watching the BIOS churn for twenty seconds on a laptop when I am constantly turning it on and off on trips--yucko wasted power.

When are we going to see something like Sun's OpenBoot as our BIOS?

Re:Do we want this?

[sklib]

Most of the time spent booting up is really the operating system's fault -- Linux and Windows are pretty bad at this. Let's not even mention the HP-UX units sitting in the computer labs someplace on this campus...

So shaving off a few seconds off the boot time is great, but it seems like the major problem is still the operating system.

Re:Do we want this?

[ncc74656]

And there's nothing wrong with that mentality either. Computers are tools. My machine takes like a minute to boot the OS... how many other tools or appliances take that long to boot? What if your TV took 60 seconds before it could perform its task?

You must never have heard of vacuum tubes. :-) OK, so the average tube TV took maybe 30 seconds or so before it got up to speed, but still...

(I speak from experience...when I was a kid (and no, I'm not that old...we're talking mid-70s to early 80s here), we had a 13" B&W TV, tube-based and probably of early-70s manufacture, in the kitchen. It took half a minute or so for it to get going.)

Even today, the radio on which I listen to Rush at home is a fifty-year-old RCA that I rebuilt a few years ago. It takes some time to warm up as well.

Re:Do we want this?

[Anonymous Coward]

If you are running Windoes98 and rebooting a PC that freezes several times a day or requires several reboots to intall a piece of software, boottime can be quite noticeable. Unfortunately Windoes startup time is most of this and not helped by this improvement.

Re:Do we want this?

[tkanerva]

Yes we should want this. Or at least remove some of that cruft that exists in current BIOSes that test the memory in the very same fashion as has happened for almost 20 years now. Not to mention all those boring messages about IDE disks connected, Press N for network boot, ad infinitum... Getting to lilo shouldn't take more than 15 seconds, linux itself will anyway reset most of the devices by its own means.

Just to have a comparison: my PowerBook G3 does the POST in about 3-4 seconds after turning it on, and then it proceeds to load the linux kernel. After Tux shows up on screen, it's 17 more long seconds until gdm greets me. IMO, every firmware (in a semi-perfect world) should be able to execute ELF binaries :-)

One of the remaining things i really dislike about x86 arch machines, is that they never took off the legacy stuff, even if the athlons have very modern instruction execution units, the systems itself make too many compromises, use outdated i8254 for cascading interrupts, etc etc. My OpenPIC supports 4 CPUs and 64 IRQ's and that's what I think how systems should be built. I don't like MacOS but I do like the way how Apple designs modern hardware.

Re:Do we want this?

[mgblst]

TV's did take a few minutes to "boot up" in the old days, and look how they turned out!

But what this does show, is an evolution of impatients in our devices, so this step may not be such a bad thing, just bring it home to out own PC's.

And i dont see why it needs to do POST tests everytime, i mean how often does something go wrong?

Re:Do we want this?

[DickBreath]

Why are the geeks so concerned about things being as fast as possible? What's an extra 5 seconds of your life?

Because every bit of performance does matter.

Case in point.

I'm not talking about time you could go get coffee while the BIOS does POST. Take that time saved to reboot, and multiply it by the number of reboots. Now suppose the blue screen code simply jumped directly into the reboot code! Like magic! A system that never blue screens! It is always up. That's innovation!

And this is not the first instance of a major advance in the evolution of computers that was made possible by speeding things up.

For those interested in doing this:

[merlin_jim]

You can find the BIOS they used here. It has to be custom-tuned, but this kit includes the code itself, so you can build the BIOS yourself. They basically disabled most of the checks and auto-configure options; no disk seeks (reasonable enough in a highly reliable system), only check the first word of every 1K memory block, no auto-configure of IDE, etc.

I've been waiting for something like this for a while! My car MP3 player takes too long to boot up... can't wait to get my hands on this. No mention of cost, but I've sent an e-mail to their contact link and will reply to this message with price if/when they get back to me.

Re:For those interested in doing this:

[Telek]

how long does your car mp3 player take to boot up that is due to POST time??

Re:For those interested in doing this:

[merlin_jim]

It's VERY random... between 8 and 15 seconds... one of these days, I'll get a website up about it... but its running on an old PII, Slackware 5.4 (a good release that supports all the hardware I need and none that I don't, that I'm very familiar with kernel hacking in)... The LILO to music time is about 3 seconds (due mostly to compiling the kernel with only the modules I need and tuning the startup scripts), but it takes 8 to 15 seconds to get there...

Re:For those interested in doing this:

[Telek]

What if you installed a sort of "sleep" mode, or "suspend" mode? What type of hardware setup are you using anyways? I just mounted a laptop in my front (removable) with a GPS unit in the pcmcia slot and kept it always-on basically, sleep mode and plugged in inside. The GPS is really cool, too bad I don't have any decent mapping software to with it (I live in Canada).

Re:For those interested in doing this:

[merlin_jim]

Well, it's an old desktop system, so a sleep or suspend mode wouldn't be supported by the BIOS...

Besides which, the idea of having my own code for the BIOS is VERY cool. I could do all sorts of things (like sending init messages to other pieces of hardware in my car, such as LCD screens, GPS units, etc.) with that...

Re:For those interested in doing this:

[stikves]

Slackware 5.4???

There has never been slackware 5.x or 6.x! It jumped from 4.x to 7.0!

(I assume you meant 4.5)

Re:For those interested in doing this:

[sfe_software]

I agree on the Car MP3 player. My BookPC runs a custom kernel on its own boot partition, so once I hit LILO I only have about 8 seconds until music. POST takes about 7 or 8 seconds by itself, so this could effectively cut my boot time in half.

Actually we can add about 5 or 6 seconds for a cold boot - Maxtor 7200 RPM drives seem to spin up kinda slow... but still, any reduction in boot time is good. It could mean I hear tunes at the end of the parking lot, instead of when I get to the traffic light up the street :)

Re:For those interested in doing this:

[Chutzpah]

Interesting... I have been thinking about a car mp3 palyer too, but I certainly wasn't planning on putting in a hard drive, the vibrations/jumping from the car would probably kill most desktop hard drives in a matter of a couple of months (especially on quebec roads). I was thinking of doing it with a boot disk/CD loading what it needs into a ramdrive then playing the mp3's off a CD (maybe make all my MP3 CD's bootable...) .

CD-ROM's also tend to initialize faster than hard drives...

Another option would be to find some way too store the OS on a flash card and a way to boot off it, flash RAM cards are getting pretty cheap nowadays with all the digital cameras around.

Re:For those interested in doing this:

[merlin_jim]

I'll check tonight... I was almost positive it was 5.4, though.

Re:For those interested in doing this:

[cloudmaster]

Mine boots up at a nice consistent 12 seconds. I'm running a Fic VA 503+ and a POS Cyrix space heater chip (it was cheap). You can gain a little time by sticking the kernel right in the boot sector, but you've gotta strip it down enough to make it fit. Did you turn off all of the BIOS auto-detect crap you could?

You might look into the Linux BIOS project at http://www.linuxbios.org/ as well. I'm sure it's mentioned farther down in teh comments here, but I haven't read that far yet. ;)

Re:For those interested in doing this:

[sfe_software]

I actually haven't had any problems, and I've had my MP3 box for over 2 years. This 40 gig Maxtor has been in several different cases, had Win98 originally (now Linux, but still FAT32 on the MP3 partition). It's been in a couple different cars, and I drive like a maniac :)

No problems at all. I even hit a deer @ 70 MPH about a year ago, and the music didn't glitch one bit...

The only thing I'd do is be sure not to mount it tight. Originally I had it in a mid-tower case which I would simply set in the trunk; the BookPC it's currently in fits under the seat, or just about anywhere. I generally just lay it in the back seat since I use it in the house as well.

Re:For those interested in doing this:

[sklib]

If you intend to use CD's anyway, you can just get a cd player that will play mp3's from a disc. Check out easybuy2000.com -- they have that stuff.
I've got an MPTrip, and it's been working very well for at least a year now.

Re:For those interested in doing this:

[Chutzpah]

Well, the problem with those is they dont play stuff like vorbis... (about half my music collection is in vorbis format now)

and if i make my own i can eventually get a nice little LCD and use it for fancy vis and stuff...

99.99999%? That's great...

[Wakko Warner]

...assuming you ever reboot. I've got machines that have been up for 300 days, and I've seen machines that have been up for 700. What good does a minutes' worth of saved booting time get me when I do it less than once a year?

- A.P.

There was no PIII/400

[sootman]

Was that supposed to be 1400 MHz PIII, or PII/400, or some other speed of PIII? (Or was it a Celeron, or 286?) I checked the article, they got it wrong there-- not /.'s fault.

Re:There was no PIII/400

[merlin_jim]

This was most likely a low-power version of the PIII... this company specializes in low-power and embedded systems, and it is common in that market to purposely underclock a CPU (or make a custom-run specially built to be underclocked) in order to save on power and heat.

Re:There was no PIII/400

[claydean]

If it is an underclocked chip they hould point out that fact. This is clearly misleading

Re:There was no PIII/400

[jeffy124]

first paragraph, second to last sentence:

The demonstration ran on an Intel SOYO motherboard with an Intel 440BX chipset and 400Mhz Pentium III processor.

You do raise an interesting point. Personnally, I dont recall P3's going that slow (P2 400 were top-of-the-line when I started college Sept '98). But nor do I recall P3's blazing along at 1.4 GHz. Maybe it's an underclocked CPU?

Re:There was no PIII/400

[claydean]

Have sent an email to General Software asking them to clarify. Will post their response if/when I recieve it.

Re:There was no PIII/400

[flatrock]

http://developer.intel.com/design/intarch/pentiumi ii/pentiumiii.htm

It shows a low power pentium III at 400 MHz.

Man...

[Syberghost]

Can you imagine a B...

...never mind.

So ...

[Rev.LoveJoy]

the obvious troll question is:

If Linux is so darned stable, why do you guys care?

yes, that was sarcasm
- RLJ

So? I've achieved faster boot times.

[Anonymous Coward]

I've personally achieved faster boot times using
Intel's System Firmware library.

http://developer.intel.com/platforms/applied/sof tw are/firmware/acsfl/abstract.htm

Straight from rom to 32bit protected mode
without passing through typical BIOS steps.

Re:So? I've achieved faster boot times.

[Anonymous Coward]

I don't think that's Free software.

How dare you post that shit here.

Re:So? I've achieved faster boot times.

[Anonymous Coward]

Actually, it is _free_ software.

It's about time...

[Anonymous Coward]

It's taken 20 years, but they finally got a system
that can boot faster than a TRS-80.

Re:It's about time...

[zulux]

I know you were talking 'bout the old Model I/II/III CoCo etc..

Did you know that the Tandy 1000 had most of MS-DOS burned into ROM - the dumb things could get you to a DOS prompt in about ten seconds from a hard boot and about 5 seconds in a soft boot. They were great for programming odd little assembly programs, cause if your program locked, you could be back to work in fifteen seconds.

Re:It's about time...

[srw]

We used to have problems with these, though. If you turned off the memory check, and installed a hard-card (Hard disk mounted on a controller card) the system would boot before the hard disk would spin up, and the HD would not be detected.

PIII 400?

[Anonymous Coward]

Ha...no such thing.

Get a "REAL" Computer

[Ziegerektum]

Everyone doing server stuff should drop thewir x86 boxes and get a Sun right now!

Seriously

The SPARC architecture does not use a BIOS with REAL mode drivers for booting. It has protected (or whatever it is in non-x86 parlance) mode drivers built right into the firmware. On x86, the BIOS contains Real mode drivers, THis was fine for operatings systems like DOS and Win 3.1. However, modern OSs (Windows, Linux, etc) need protected mode drivers. BY placing these right in the fimware Sun is able to smoke x86 performancewise ALWAYS. I thin its time to ditch our legacy DOS hardware and start getting x86 machines with protected mode BIOS drivers. Anyone with more technical information, please comment.

Re:Get a "REAL" Computer

[Anonymous Coward]

Not only that, but Sun's new sunblade machines retail for less than $1000 for the basic model, and can run windows concurrently with Solaris. It's time to abandon lame-ass x86 machines.

Re:Get a "REAL" Computer

[Ex Machina]

I have no idea why this is modded down to zero.

These are some of the reasons Commercial UNIX filesystem performace has a slight advantage over FreeBSD/Linux on x86. It'd be really nice to see some legacy free PC hardware!

Re:Get a "REAL" Computer

[Anonymous Coward]

but I need to run my old 16 bit real-mode DOS programs.

Re:Get a "REAL" Computer

[throx]

It's a bootstrap. It's going to have different drivers from the operating system anyhow, so what does it matter if they are real mode, protected mode or a-la-mode. There is no significant speed difference between real and protected mode anyhow (it's running on the same CPU) so I think your point here is way off the mark.

I really don't think everyone is going to drop x86 and by SPARC because of boot times. x86 has always blown SPARC out of the water in price/performance anyhow, and what am I going to do with all my x86 software if I move?

Re:Get a "REAL" Computer

[throx]

Given that the Blades that retain for under $1000 have less than a third of the performance of a similarly priced x86 machine, can you explain to me why on earth I'd really want one of these?

Re:Get a "REAL" Computer

[rootmonkey]

Cheap hardware clusters work fine and still protect against systems when server(s) go down, and they will. x86 is great for systems that can be clustered. I'd rather have several inexpensive servers that have less power than one super server, because one super server is still one.

Re:Get a "REAL" Computer

[Anonymous Coward]

Performance's not everything.

Soon after I got my Blade I sold my PC. Blade is QUIET (no CPU fan), looks great (I hate beige plastic tower cases) and has all the grunt I need at home. Solaris 8 is great too.

Re:Get a "REAL" Computer

[flakac]

I thought the SPARC bios uses interpreted Forth...

Re:Get a "REAL" Computer

[djp928]

Because people who know anything know that looking at a MHz to MHz comparison across different architectures is meaningless? Hasn't your AMD bias tought you anything?

Clock for clock, the UltraSparc architecture blows away anything in the x86 world.

Re:Get a "REAL" Computer

[idiot900]

No modern OS on x86 uses BIOS calls to access the hardware anymore, so the only other reason to have your computer in protected mode 100% of the time is to impress the ladies.

Re:Get a "REAL" Computer

[The_Messenger]

I have one of these machines... it makes a great little development server. You're right; I could have built a faster x86 box for less... and indeed I have in the past. So why did I buy the Blade 100?

Three words: Sun hardware fetish. :-)

Re:Get a "REAL" Computer

[marxmarv]

Anyone with more technical information, please comment.

Considering that almost all your information is wrong, I'll be happy to correct it as best I can.

The SPARC architecture does not use a BIOS with REAL mode drivers for booting.

The SPARC doesn't have or need real mode.

It has protected (or whatever it is in non-x86 parlance) mode drivers built right into the firmware.

No it doesn't. It has drivers written mostly in Forth bytecode, just capable enough to bootstrap the OS. See also: OpenBoot.

However, modern OSs (Windows, Linux, etc) need protected mode drivers

Drivers do not operate in a vacuum. Among other things, they need to initialize/uninitialize themselves, manage (allocate/free, map/unmap, lock/unlock) memory, cooperate with other drivers, and share various resources in the fashion that the host OS requires. None of that has squat to do with what addressing mode the processor is running in and everything to do with what OS is running, which is why OS'es come with their own drivers. See also: UDDI.

BY placing these right in the fimware Sun is able to smoke x86 performancewise ALWAYS.

Bullshit, bullshit, bullshit. The weaknesses of the IA-32 architecture descend mostly from 20 years of backward compatibility for marketing purposes, and the resultant need to handle legacy crap from bit-banging 1980's programs and drivers to 16-bit-addressing DMA controllers to slow ISA peripherals to IDE controllers that you could still run on a PC/XT ferchrissake. More function = more silicon = longer critical path = slower.

I thin its time to ditch our legacy DOS hardware

Yes...

and start getting x86 machines with protected mode BIOS drivers.

The BIOS is history once any modern OS boots (with the possible exception of power management on laptops). What's in the bootstrap ROMs doesn't fscking matter once the OS is loaded.

The major problem is the chipsets and the 20-year-old designs they're based around. Drop in full 32-or-more-bit DMA controllers or require all peripherals be bus-master capable, segregate the ISA bus to its own out-of-the-way 16MB window somewhere (see: Apollo DNx000 family), hardwire a handful of interrupts and a hardcoded address range to each slot (see: EISA), drop the legacy keyboard/mouse interface, and redo IDE entirely (see: SCSI). While we're at it, let's scrap BIOS and replace it with OpenBoot. Now there's a machine free of legacy crap that might be worth writing home about.

-jhp

Re:Get a "REAL" Computer

[SumDeusExMachina]

Blade is QUIET (no CPU fan), looks great (I hate beige plastic tower cases)

Well, whooptie shit, it matches your furniture. I suppose you bought a Mac too?

has all the grunt I need at home.

Yeah, and for twice the price, how can you go wrong?

Solaris 8 is great too.

If you have a fetish for slow OSes that only work well on massively MP platforms, then yes, Slowaris is great.

Re:Get a "REAL" Computer

[Anonymous Coward]

Well, whooptie shit, it matches your furniture. I suppose you bought a Mac too?

Ha-ha.

I perfer to have my computers QUIET, tidy and practical. Not some ugly pile of bare metal that sounds like a turbine engine.

You seem to be measuring dick-lengths here. Ok, you've got a speed fetish and are prepared to pay to get it. I on the other hand like the things I mentioned above and I am prepared to pay to get them. To you it might seem silly, but to me your need for speed is silly too.

Re:Get a "REAL" Computer

[Anonymous Coward]

AWESOME! my troll gets crapflooded!

Re:Get a "REAL" Computer

[Anonymous Coward]

don't forget about the pile of SHIT that you get when the COCK fan fails (most systems will now give off alarm farts when COCK temperature rises to high, but that doesn't help you when you are working a nice attractive ass and don't have time to react to farts coming from the rectal processing unit).

Re:Get a "REAL" Computer

[Urchlay]

Well, if you're running *nix on x86, just recompile it for Sparc :)

As far as price/performance goes... try looking for a used SparcStation 10 or 20 on Ebay, I've found a few for around $100, which outperform any $100 x86 PC you could likely buy... unless you get a really amazing deal on a used PII or better machine with SCSI disks...

Yes, I realize you are talking about the new computer market, not used... But even there, will a brand new $40,000 Sparc machine (Enterprise server or whatever) outperform a $40,000 x86 machine? (Does anybody sell $40,000 x86 machines??)

But these are extreme cases anyway... in the real world, you're right.

Re:Get a "REAL" Computer

[glassware]

BY placing these right in the fimware Sun is able to smoke x86 performancewise ALWAYS.

While I agree that Sun's computer architecture is more clean than that of the PC, SPARC CPU performance really isn't that good. Using the best benchmark (SPEC CINT 2000) with which I am familiar, an Athlon 1.4GHz can get up to 495, while the fastest Sun computer (a Sun Blade 1000 model 1900) gets a respectable 439. A Pentium 4 1.8GHz can score as high as 596.

While CINT2000 is a great measure of typical application performance, it used to be true that Intel PCs had terrible floating point performance compared to RISC architectures such as Sun, SGI, and Alpha. However, according to SPEC CFP2000, the Sun Blade scores 369, the Athlon scores 433, and the P4 goes up to a shocking 618.

Sadly, most of today's software isn't CPU limited, so your best choice is to get a balance of performance between your disk, memory, operating system, graphics card, and CPU.

Re:Get a "REAL" Computer

[Nexx]

Does anybody sell $40,000 x86 machines??

Dell/Compaq/Gateway/IBM/Me would most happily sell you a single desktop PIII for $40,000 :-)

Seriously, depends on what you consider a part of the machine and what you consider a part of the storage "mesh", and you probably would be able to configure a $40k machine. Not to say it will outperform (depending on the performance criteria) a $40k Alpha (RIP) or a Sun, of course, but you probably could do it.

Re:Get a "REAL" Computer

[throx]

As far as price/performance goes... try looking for a used SparcStation 10 or 20 on Ebay, I've found a few for around $100, which outperform any $100 x86 PC you could likely buy... unless you get a really amazing deal on a used PII or better machine with SCSI disks...

Oh, you wanna talk USED machines? Sure, I can beat that on price/performance. I got an old K6 box for free the other day because no one wanted to buy it. Hard to beat that with any $100 machine. ;-)

As for $40,000 machines - hell yeah! Look at the TPC results at www.tpc.org. Most of the x86 machines there are worth in excess of $500,000.

Re:Get a "REAL" Computer

[Anonymous Coward]

Not to mention that 32-bit BIOSes have been around for over 5 years now, and do contain protected-mode BIOS drivers.

Re:Get a "REAL" Computer

[redzebra]

Yep.. your defeatly right on all points and openboot would be the real neat replacement.

just wanted to add that I don't even see why
protected mode drivers would be any faster than real mode ones.. i would guess the bios is mostly doing polling/io stuff also full 32 bit instructions are still available in real mode if
they need fast moves

--red

Re:Get a "REAL" Computer

[Anonymous Coward]

It has protected (or whatever it is in non-x86 parlance)

Actually, there is no equivalent term for "real mode" on SPARC or any other modern (or semi-modern (i.e. 1979 or later)) processor family. Only x86 is "gifted" with this oddity of evolution, because it's the only family that would need it.

(Remember: just because evolution happens in Real Life, that doesn't mean it's not a Bad Thing.)

Re:Get a "REAL" Computer

[rew]

Sun has forth code in the firmware. That's MUCH slower than the 16 bit legacy we have for DOS.

Roger.

Re:Get a "REAL" Computer

[The Paradox]

However, modern OSs (Windows, Linux, etc) need protected mode drivers

Won't disagree with the speed considerations here, but I just thought I'd point out that no, Windows does not have to have protected mode drivers to run.

I've got a P180 from '94 that I stuck a giant hard drive in and put 98SE on, as a backup location...couldn't get the ancient, proprietary CD to work, so I'm now using real-mode CD drivers.

Older, yes, slower, yes, but it gets the freakin' CDROM working, and that's what counts on a machine like that.

Re:Get a "REAL" Computer

[maggard]

It'd be really nice to see some legacy free PC hardware!

There is - it's called a Mac. Open firmware, PPC, clean design, these days essentially a consumer-cost workstation-class box. Even runs a modern OS now (oh wait, you already know all about that.)

Re:Understanding Slashcode!

[ethereal]

That explains a lot about how you can hit the lameness filter when just cutting-and-pasting from another web page, which with Netscape will insert a lot of spaces. Good compression == lame post, apparently.

How about posting your example alternative code somewhere else, and just not licensing it so that it could be used with slashcode? It would be interesting to see if you could come up with a method that would penalize (no pun intended, really) ascii art posts but not cut-and-paste quotes.

Great

[Sheldon_Brown]

So, basically, if Windows can get down to one reboot per year, they could actually get those 5 9's they're always talking about in their ads.

I'm only crashing once a week on average under Windows 2000, so I'd say they're well on their way to achieving this by 2045. Good luck guys!

Re:Great

[Anonymous Coward]

You'll be lucky to not exceed 99999 reboots a year on a windows.

I think the 5 nine's figure is what you would get on a pc if you do not install anything other than the OS which is pretty useless.

Re:Great

[Anonymous Coward]

Well, MS can still do 100 reboots per year with that thing - MS counts 99.999 as 5 nines, while everybody else takes the ninetynine for granted and just counts after the decimal point.

Re:Great

[Amazing Quantum Man]

I took a class in WDM drivers (yes, I know...) from a guy who had asked MS about their 5 9's claim.

It's for Win2K Datacenter Server, running on Certified Hardware with Certified Drivers running Certified Apps only. Put ANYTHING that ain't certified on it, and MS won't give you that 5 9's guarantee.

In other words, you'd better be running Win2K, IIS, MSSQL, etc... forget any non-MS stuff.

I think this is kind of bogus advertising, but if you think about it, that's the way Big Iron(tm) ran back in the olden days.

just quicker to the inevitable...

[Anonymous Coward]

What good is a fast boot time, when you hit the brick wall at the end...

"my system can go from boot to blue screen of death in 3 seconds, biatch"

fear.

Something similar?

[MentlFlos]

Isn't there something like this where they throw a bootloader or something right in the flash on the mobo?

I vaguely remember this from /.

could just be that I have yet to have my coffee

-paul
meatbarn.com

Re:Something similar?

[mini me]

They have Linux running as the BIOS, this is likely what you are thinking of.

.8 sec... SO?

[xanadu-xtroot.com]

OK, while that's a pretty nice thing, what's the big deal? That .8 sec is only [button push] to the lilo prompt. So? The box STILL has to boot. What if you've got a box that still has to fire up a bunch of daemons before it's even online and usable? What if it was a dirty shutdown (and the silly fool is still running ext2) with a nGig drive(s)? How does this help uptime?

Geek

[TheLoneCabbage]

Wow! Geek perfection! Cold to LILO in 0.8 seconds. Women will flock to me!

Now if I can just get LILO working again...

Re:Geek

[Anonymous Coward]

Try GRUB. It's never had any of the nasty limitations or quirks that LILO has.

Re:Geek

[passion]

Yeah, all I can ever get is LI..

and not LAID...

Re:Geek

[Dwonis]

Do you know of a half-decent GRUB tutorial? I can never seem to collect the time to sit down and learn GRUB like I did CVS... from the manpages.

just watch...

[GaylordFucker]

any day now Intel will have a callback cause those idiots probably forgot something... i won't be surprised... better let AMD handle processors... after all they're efficient, reliable and quite stable... notice how all the big companies blow it out the rear?

Great for embedded devices, but not for other...

[Doctor_D]

Quick boot time for embedded devices is a nice thing. The last thing you want is to have your microwave take 15 minutes to boot.

Of course on stuff that matters, boot time is quite longer. Let's take for example a Sun Enterprise 10000, it takes at minimum of 20 minutes and up to a few hours just to run POST. This still doesn't include time to boot the operating system. But once that system is up and running, it stays there until either something bad happens or the admin shuts it down.

For obvious reasons you wouldn't install this BIOS in your home PC or linux server, where checking memory, detecting SCSI and drives and so on are pretty much necessary.

Win2k

[Simm0]

Install Win2k and it will take 120.8+ seconds to boot up :P

Re:Win2k

[Anonymous Coward]

Yes, that's what I did. I was sick of the 300.8+ seconds I was getting from Linux.

This is not for desktop systems!

[merlin_jim]

Okay, for everyone alternatively complaining that this is overkill on the desktop, or that they would prefer all the checks, etc. in place... this is NOT built for desktop systems.

Read the post; this is for an embedded system requiring seven nines. Though it can (and most likely will) be adapted for desktops, any desktop running this will be a high-reliability server, with all the checks (except memory, which this chip does after a fashion) built into the hardware...

Re:99.99999%? That's great...

[Anonymous Coward]

you're fat and stupid..

the point is that if a machine goes down, it will reboot right away, exactely what is mentioned.

and yes, i know what you look like in real life, and you are a fat and stupid tool. you smell terrible and i know you read this whole post because you feel cool now that you know that somebody knows you.

slashdot

[Anonymous Coward]

Slashdot also has seven-nine uptime. Except it's not 99.99999%, it's 9.999999%.

Re:slashdot

[ralmeida]

lol :)

Re:slashdot

[ers81239]

rofl :)

Re:slashdot

[Anonymous Coward]

FOAD :)

Re:slashdot

[Anonymous Coward]

ROFFLEMAYO

Re:slashdot

[Anonymous Coward]

Rollin' On The Fucking Floor Laffin' My Eye Out?

Re:99.99999%? That's great...

[hajibaba]

Once I got a call from a plant that had an old Novell server that needed to be cleaned out. The company produced cotton yarn, so this was a really dusty environment. They had an old Compaq 486DX2/66 server that had an uptime of almost 4 years! I wish I had my camera when I saw that. Try that with NT!

Faster booting systems

[tshoppa]

My Atari 2600 boots and is running the "application" in about a tenth of a second :-).

While a fast boot is obviously good for those embedded developers who are stuck with PC-clone hardware, I think the lesson here is that the PC-clone architecture is only barely acceptable in many embedded real-time situations.

Re:Faster booting systems

[Wavicle]

Does that include the time blowing on the card edge or dragging an eraser across it hoping to get good enough contact to boot?

Ahh, the memories.

Re:Great for embedded devices, but not for other..

[baptiste]

Quick boot time for embedded devices is a nice thing. The last thing you want is to have your microwave take 15 minutes to boot.

The last thing I want is my Microwave running a Pent anything and some variant of Windoze. When that happens I'll be out back with some wood, rock, and flint to cook my meals :)

Re:99.99999%? That's great...

[smack_attack]

Just imagine this in an environment where you can't keep the machine running all the time... like your car.

About time - was able to do this _years_ ago

[ehud42]

My ][e could boot to a command prompt in less then 1 second. A fast (by historical standards) HD meant a boot to a DOS was easily only a couple of seconds.

(This is probably going to be flagged as a troll or flamebait, but think about it. We have put up with crap for so long, that when we finally get sub-second boot times back it's a big deal. It's like hiding toys from your kids for 6 months and then bringing them out as winter sets in - they get all excited about stuff they used to have.)

I'll admit ignorance as to all the required checks, double checks and initialization that must go on to get a decent OS up and running, but I still can't help but think that inefficiently designed / written bloat-ware could be done much better to improve the boot times of modern machines. Why not lazy load the drivers, etc as required?

Re:About time - was able to do this _years_ ago

[Anonymous Coward]

It's only the BIOS boot time. The OS still takes an hour to boot up.

Re:About time - was able to do this _years_ ago

[stikves]

Actually linux boots in less than one second. If you ever looked at the boot logs, you'll see the line

INIT version xxx booting...

This means that OS is ready at that point but is starts loading support programs (daemons, etc). If you ever modify /etc/lilo.conf to add [append = "init=/bin/sh"], you'll see what i mean.

The the time after 0.8 + 1.0 secs is occupied by scripts. The scripts does three things (mainly):

• Loading modules
  
• Loading daemons
  
• Configuring software and hardware environment.
  

So if you were to make a MP3 player, you can modify init=/bin/mp3boot.sh which contains:

#!/bin/sh
/sbin/depmod -a
/sbin/modprobe emu10k1
/sbin/mount /usr /mp3
/usr/bin/mp3123 /mp3/*

this will boot in 4 secs with that BIOS

Re:99.99999%? That's great...

[Anonymous Coward]

Ocassionaly certain devices(telephone switches) require a reboot for firmware upgrades. With this device the reboot time can be minimized thus, maintaining HIGH-availability required in this environment.

We're not talking about crashes here. Crashes are simply out of the question.

Can your Linux box do that?

Yes

[bluephone]

These baords/BIOSes aren't meant for J. Random Hacker. They're meant for the embeddable computing sector. These types of systems don't change the hardware setup, so all that configurability is useless. These are systems where downtime means something BAD. Thinks nuclear plant regulator systems, telephone switching systems, etc. These could even be used for high-end PBX style boxes where there's zero need for constant upgrades and hardware swapping, but you need fast off-to-operational status.

Re:.8 sec... SO?

[praedor]

What I'd like to see is an EPROM to contain a kernel. Any kernel. A generic 8 megabyte EPROM that will hold the linux kernel (or, if you are stupid, a doze kernel) and various modules so that you can very quickly bring up your system instead of waiting for this daemon, that daemon, etc.

Sure, some of this stuff would still have to take place outside the EPROM, delaying powerup-to-input time but it could be minimized by sticking as much as the user wants (and can fit) into the EPROM. It should be generic so that it can hold doze, linux, sunos, freebsd, whatever you want as your primary os.

My system remains much the same from boot to boot so I wouldn't need to constantly reprogram the EPROM to fit with my latest change. There could be a simple utility like a BIOS upgrader app to handle the EPROM programming. Make it so that it isn't absolutely required so that if something goes wrong with the EPROM you can still boot off you harddrive - which you would need anyway if you use a bootloader and want to bootup another os periodically.

What is it that prevents this sort of thing? Design mobos with a new chipslot for the kernel EPROM and design the chip to contain enough mem to hold any one of the os kernels that are generally used (or likely to be used). If there is a lot of leftover space, perhaps you could fit another kernel and supporting modules into it until it is full allowing for a very fast dual boot setup.

Re:Great for embedded devices, but not for other..

[Cock Knocker]

You don't think it would be cool to have a neural network that adjusts output based on sensor readings of what its cooking and how done it is?

That's going to take some number crunching. Granted you could do that purely in hardware, but there's a reason we make the hardware generic and wicked fast & write software.
You sound like the people who say that there's no reason for anyone to buy anything over 1GHz because there isn't any software to take advantage of it - such people
1. Have no imagination, and
2. Are content eating rubbery chicken and half popped popcorn (and the software equivalents).

ka-pow!

Just boot the kernel directly!

[Big Dave Diode]

You want fast boots? Put the KERNEL in the BIOS. No need to run BIOS code if you don't have to.

Re:Just boot the kernel directly!

[mrnobo1024]

The BIOS is in memory addresses 0xF0000-0xFFFFF, which is only 64k. The Linux kernel is way bigger than that. How can you fit it in?

Re:Just boot the kernel directly!

[orionpi]

Just put a loader part of the kernel there the rest at another address.

Re:Just boot the kernel directly!

[Anonymous Coward]

The current BIOS chips are bigger than 64K, the rest is paged in on demand.

Fast Boot Commodore

[Haxx]

My Commodore 64 boots in .1 seconds

soyo and 7 nines

[bziman]

The embedded system was clocked at 0.8 seconds from system power-on to transfer of control to LILO. This was on an Intel SOYO motherboard (440BX chipset) running a PIII 400

Two comments:

Doesn't Soyo mean "gentle" in Japanese?

And, if you really need "7 nines" uptime, you shouldn't be relying on a single processor -- you should be relying on a processor farm that supports hot-swappable processors, so you can lose one or two or fifty and only lose a fraction of performance for as long as it takes to replace those processors.

On a high availability machine, there should never be any reason to reboot until you must upgrade the kernel, and I'm sure there are ways to do that without requiring a hard reboot. IBM has had farms like this for years.

--brian

Re:soyo and 7 nines

[pne]

Doesn't Soyo mean "gentle" in Japanese?

Depends on how it's spelled. apparently means "elementary attainments"; is "itch"; and is "clothing". I'm not sure how it means "gentle", unless you mean the "soyo" of "soyokaze", which seems to mean "gentle breeze" -- but I don't know whether that soyo appears in other contexts (and is not usually read "soyo").

But if I understand correctly, Soyo is a Taiwanese company whose Chinese name is (read mei2jie2 and meaning maybe something like "victorious plum"). http://www.soyo.com.tw/chinese/com-profile.htm gives their company profile in Chinese and appears to explain where the word SOYO comes from; unfortunately, I don't know enough Chinese to understand what they are saying.

Cheers,
Philip.

Re:soyo and 7 nines

[The_Messenger]

This is what SUNs do best. On an Enterprise box, you can add and remove CPUs and RAM modules while the machine is running. Very few shops need this kind of uptime, but having seen it done, I can say that it's very impressive.

At the other end of the scale, we have the PC. I once killed a CD-ROM by trying to connect the IDE power while the machine was on. :-)

Re:soyo and 7 nines

[wurp]

As has been pointed out several times, this is for embedded systems. I don't want to have a processor farm with hot-swappable processors controlling my car's engine, brakes, steering, or my doctor's medical equipment.

Of course, I didn't catch that until I read posts pointing it out, and you probably have caught it by now too, but I couldn't resist ;)

Do the math once more.

[suso]

Seven nines of uptime is 315 seconds a year. I think you mean nine nines of uptime. Or maybe you mean seven nines after the decimal.

Re:Do the math once more.

[mrnobo1024]

99.99999% uptime = 3 seconds downtime per year
99.9999% uptime = 31 seconds downtime per year
99.999% uptime = 5 minutes downtime per year
99.99% uptime = 52 minutes downtime per year
99.9% uptime = 8 hours downtime per year

Re:Do the math once more.

[Quietust]

Perhaps you forgot to divide by 100...
1 year ~= 31556926 seconds (according to a nearby TI-85).

99.99999% uptime == 0.00001% downtime.
0.00001% == 0.0000001
31556926 seconds * 0.0000001 ~= 3.16 seconds.

Re:Do the math once more.

[JediTrainer]

Seven nines of uptime is 315 seconds a year. I think you mean nine nines of uptime. Or maybe you mean seven nines after the decimal.

Hmm...

60 seconds * 60 minutes * 24 hours * 365 days = 31536000

99.99999% of that is (0.9999999 * 31536000) = 31535996.8464

31536000 - 3153996.8464 = 3.1536 seconds

Or the quicker way: (60*60*24*365 * 0.0000001 = 3.1536)

Looks like you forgot your decimal place somewhere.

Re:Do the math once more.

[mattyohe]

That TI-85 must be fucked up then.. there are 525600 minutes in a year. 525600*60=31536000 but obviously still ~ 3.16 seconds. now the question.. can the services and other processes be loaded in the remaining 2.36 seconds?

Re:Do the math once more.

[mrnobo1024]

Actually, there are 31556926.08 seconds in a year. (60 * 60 * 24 * 365.2422 = 31556926.08)

Re:Do the math once more.

[mattyohe]

fine we are down to symantecs :)

Re:Do the math once more.

[Quietust]

Actually, that TI-85's definition of a year is about 365.24 days. Leap-years, you know...

Re:Do the math once more.

[mattyohe]

Thats why he said "Actually, there are 31556926.08 seconds in a year. (60 * 60 * 24 * 365.2422 = 31556926.08)"

Re:Do the math once more.

[Anonymous Coward]

Notice the time difference between his reply and mine.
About 9 minutes.
A lot can happen in 9 minutes, like starting a reply, typing a bit, punching some buttons on a calculator, answering a phone call, then submitting the reply.

-- Quietust, posting anonymously because this message does not pertain to the article

Re:Do the math once more.

[Amazing Quantum Man]

fine we are down to symantecs :)

So Peter Norton gets to determine how long the year is?

Ohhhh... you meant semantics!

This computer is sooo fast

[sniglet999]

"I can reboot twice as much in half the time!"

--Direct quote using a p60 around 1994

So how many times can I reboot in .8 seconds and still maintain 99.99999%? Or rather, why don't you just cluster your services and reboot the individual servers whenever you want?

Re:99.99999%? That's great...

[CrazyBrett]

Eh, that's nothing! There's a company in Europe somewhere (I forget the name) that's been running an OpenVMS system for 17 years with zero downtime. Now that's impressive :)

-- Brett

Heh...

[Anonymous Coward]

Using my well-shod foot, I can boot a Pentium MUCH faster than that!!

I don't know about you...

[GiorgioG]

My system spends more time loading the OS than the BIOS cycle - why does anyone really care?

Find me a stable/decent desktop OS that boots in less than 5 seconds like a gamebox - and I'm ready to jump ship ;-)

practical uses

[psych031337]

This is very clearly a product aimed towards users of 9x versions of the popular Redmond based OS. No one else will get a kick or efficiency boost out of (re-)booting that fast.

If it's been said before, well d'uh. I ain't got no karma to lose.

Re:99.99999%? That's great...

[Anonymous Coward]

IIRC that's the national train service in Ireland

Intel names are all about marketing hype.

[stinkgeek.com]

The first PIII was a 450. Regardless, the biggest difference between the PII and III was the name. Intel released the faster models under the name PIII but it was basically a PII with a slightly larger cache. That is why you could use it with the same motherboards. And Celerons are basically what the PII used to be, they wanted to "cheapen" the name of the low end processors or else they might hurt PIII sales. Just like the PIII will soon be renamed Celeron in order not to hurt the sales of the P4.

Raises an interesting question...

[M_Talon]

Ok, you turn off a majority of the system checks to make this possible. Nice if you plan on having a regular reboot schedule. However, if the system fails due to hardware, you're forced to go in and turn all the checks on, then reboot again just to find out what broke. Seems to violate that "seven nines" rule, doesn't it? You'd have to have a lot of faith in your system being reliable to do that.

There is a PIII/400

[alienmole]

It's a mobile Pentium III.

Here's the URL spelled out in full to get past the $&#$(! filter: http://www.intc.com/pressroom/archive/releases/dp1 02599.htm

Gee Wizz

[Anonymous Coward]

da Amiga 4000 took 13 secs from power on to HD stop, full GUI. Shutdown was 0 secs(power off). OK so you would have to do it more timnes that you would want to. Is there another GUI platform that comes close? or is there a way of getting Linux or windoze to perhaps load completly from a memory image?

Re:Great for embedded devices, but not for other..

[baptiste]

You sound like the people who say that there's no reason for anyone to buy anything over 1GHz because there isn't any software to take advantage of it

What I may sound like and what I am are two totally different things. I've already got a 1GHz server in my house and bought a 700MHz AMD Athlon when they first came out. I tend to be an early adaopter. But even so, there are some things that , to me, are overkill. And Pentium III processors running non Real-time OSes in appliances just don't get me excited. No thanks. Embedded devices have to start fast but they also have to be super stable. In teh embedded arena - simple is best to avoid getting burned later (I do embedded design) I'm perfectly happy with my Microwave - its 10 years old and works just fine. My chicken isn't rubbery cause I don't cook in my Microwave. I have an oven, stove, and Weber grill to do that. My microwave is for reheating leftoves and well, leftovers never taste like they did when tehy were cooked.

The point here is that just because you think some things are overkill doesn't mean you have no imagination. Hell, I've designed home automation gear - and I still think the idea of networking all your appliances together is looney. Scanning items in my fridge? Yeah - that's gonna happen - talk about a waste of time! But hooking up an central controller to your house systems (HVAC So don't be so quick to judge next time. I'll probably buy the first 0.13 micron Athlon that comes out. But a Microwave running a PC based controller - gack - no thanks.

It's closer to 9 7s uptime

[wiredog]

77.7777777% uptime is better than 9.999999%

Re:It's closer to 9 7s uptime

[Anonymous Coward]

You really didn't understand the joke, did you.

Re:99.99999%? That's great...

[CyberKnet]

Over Y2K the company I worked for insisted EVERYTHING be turned off and physically unplugged from any network point, power point and phone point.

The cause for their worry wasnt a Y2K glitch in any of those sub-systems. They were worried mostly about a power surge or spike when everyone who had turned off their equiptment over Y2K turned it back on again just after midnight.

Of concern to us was an old HPUX system. We just werent sure if we could reliably turn it off and then get it going again. The reason?

It hadnt been rebooted in 10 (TEN) YEARS.

Re:99.99999%? That's great...

[DeanT]

Ocassionaly certain devices(telephone switches) require a reboot for firmware upgrades.

I seem to remember telco switches run some special hardware that amounts to a computer with two system planars. The idea was to let them have a failure or do upgrades without taking the switches off-line at all. Of course, my memory could be a little fuzzy since I haven't had my 27 cups of coffee yet... ;)

DeanT

LILO?

[germinatoras]

Sounds like it'd be much more useful for Windows than anything that LILO normally handles. I guess maybe for kernel re-compiles and such, but with kernel modules and most everything else runing in user-space, I almost never reboot my Linux box.

Who cares?

[Pig Hogger]

Who cares if you take .8 seconds to LILO? You don't have to boot a Linux machine 32,767 times a day anyway!!!!

Linux Variant with Zero Boot Times

[StCredZero]

EROS is an operating system that doesn't boot. Instead, it just loads a memory image from disk, which can be lightning fast when it's arranged as one contiguious file to eliminate seek times.

Re:.8 sec... SO?

[Telek]

you think that you can fit everything that you need for a linux or windows bootup in 8MB of flash? You think that you can fit BOTH of them?

The linux kernel is small, yes, but that's because all of the needed modules and drivers aren't in it! They're loaded on the hard drive.

Not to mention that flash is very slow... and expensive...

You'd be better off to store a memory image of a booted kernel at the beginning of your hard drive, along with all necessary information to initialize all of the hardware. Just have a small bootstrap/lilo type of thing that quickly loads up enough to access the hard drive and file system, then load the rest into memory directly, then initialize the hardware.

But I reboot so infrequently that it doesn't really matter how long it takes. Hell, I have my system set to do a full memory check on bootup. It takes an extra 45 seconds, so what?

And stop bashing windows... My W2K Server has been up for 145 days now and counting. Check it out along with CodeRedII attack info realtime (yeah, shameless plug) =>

Re:.8 sec... SO?

[xanadu-xtroot.com]

Then I suggest you look into this page that is working on a Linux BIOS.

Enjoy.

Will it matter?

[edox.]

Evergreen's Performa 500 can boost your early Pentium II to 500-MHz Celeron power--but it's costly Early adopters of the Pentium II processor were left in the lurch when Intel raised the bus speed for 350-MHz and faster Pentium II systems from 66 MHz to 100 MHz. Stuck with the older motherboards and chip sets, those users could not take advantage of faster processors. Enter Evergreen Technologies into the breach, offering its Performa 500 upgrade module, which features a 500-MHz Celeron CPU that runs on the 66-MHz bus. The Performa makes for a clean processor upgrade, since you won't have to buy and install the extra parts that assembling your own Celeron CPU upgrade requires. But there's a heavy premium on that convenience--the Performa's $299 price is a bit steep for what you get. Click here for more..

I don't think downtime/year is the key issue.

[SagSaw]

Rather, I think the issue is minimizing the unexpected downtime that occurs at a critical moment. Lets say you are using a computer in surgical equipment. Let also say that, heaven forbid, there is a bug in the software code. While the surgeon is busy fiddling around inside the patients head, the equipment freezes up. Every second that it takes to restart the equipment, there is the possibility of harming the patient.

Obviously, equipment this critical should never ever crash. It's nice to know, though, that should something happen, the equipment will restart quickly.

fallacy...

[Telek]

99.99999% uptime per year?

3.15s/year downtime.

This is possible with a reboot? of any machine? Even if the post takes 0.8 seconds?

I think the only way to have that type of uptime is to not go down at all.

Or of course, you could say that you have a 99.99999% uptime (average, over 1000 years) ;)

Re:fallacy...

[merlin_jim]

Actually, the machines running these are likely to have a very trimmed kernel, specifically built for this kind of thing...

Forget the seven nines thing, point is, this is a BIOS for a system that has to be up and up all the time. If it ever needs to reboot, it can't stand a reboot of more than a couple seconds... and for whatever reason, you can't have redundant processors (think in-car computer)

Re:fallacy...

[stikves]

Yes it's possible. Read my comment above...

Now if PS2 could do this...

[Uttles]

I would consider buying one instead of waiting for GameCube... but I'd probably still go with GameCube.

Should we be surprised?

[nobodyman]

From the website: The adaptation included disabling memory count-up, banners, and various POST status messages; using the "quick" memory test, which scans only the first word of each 1KB unit of memory during the memory test; and disabling floppy and hard disk seeks during POST.

Disabling the memory count and disk seek are options in most BIOS's. Won't these make *any* mobo boot faster? Can't say I've ever tried.. Win2k still takes two minutes (and what's with the three seperate progress bars?)

--nobodyman

Boot in seconds, millions of them...

[iomud]

Unfortunately lilo is set to prompt with no timeout, it sure gets there fast though.

Re:.8 sec... SO?

[SuperguyA1]

You're missing the point.
In order to be able to claim %99.99999 uptime you
would have to do your reboot in 3 seconds. It is
possible to boot in 2.1 seconds in a high
performance box with a tight kernel. If the person
running the box is 'a silly fool still running ext2, then they probably don't care about the
%99.99999 uptime.

What about boot down?

[jmerelo]

OK, it takes 0.8 seconds booting up, but what about booting down? It probably takes much more (killing processes and all), which means that, unless "powering down" means "unplugging", it'll take a bit more to reboot the whole machine; I guess the relevant figure for uptime should be "reboot" time, not just "boot" time.

Re:What about boot down?

[Anonymous Coward]

Just pull the plug. 0 second boot down.

Re:What about boot down?

[jmerelo]

But then boot up will take hours while the system is reconstructed, fsck'd and all the rest

Fast Boot Commodore et al

[Anonymous Coward]

Your Commodore64/Vic20/PETs, your Apples, your Atari 400/800/1200s etc all had their OS built into ROM (or ROM Cartridge). This is not a fair comparison. I'm sure if the Power On Self Test of even a P75 merely checked 64K of memory, initialized a few counters/buffers, and jumped to a ROM based OS, you'd wind up with a boot time in the millisecond range.

Re:Fast Boot Commodore et al

[Anonymous Coward]

Oh, and a BIOS isn't built into a ROM?! The comparison is more than far. Booting from a ROM vs. finishing the BIOS initialization tasks are roughly equivalent.

Why?

[Anonymous Coward]

Why would an embedded system where reboot time was important have the bios delivering control to LILO?

Why would precious reboot time be devoted to a dual-boot kludge?

PIII 400MHz?

[Anonymous Coward]

I think someone is tripping.... I remember the PIII entering the market at 450MHz. Am I wrong about this? I looked on pricewatch, and sure enough 450 is the starting speed. http://www.pricewatch.com/menus/m3.htm

stinking devices

[lildogie]

for a device that must support "seven nines"

(my emphasis)

A pet peeve of mine is that PHB's think that "device" uptime is the same as "system" uptime.

Decades ago, we had fault tolerant systems that had large-chunk redundancy. An entire mainframe could fail and the system kept serving.

OTOH, haven't you ever had a failing app take down your system, while running on perfectly healthy hardware?

The reason this misconception, that perfect-hardware==perfect-uptime, frustrates me, is that the PHB's get sold this bill of goods by hardware salespeople. Then they don't even allow for downtime to upgrade the effing OS every two years. Nor do they allow for a second system to either (a) take the load during an upgrade, or (b) test updates to the application.

For this silly reason, giant, fault-tolerant boxes are hurting, rather than helping, high-availability computing. Bosses would rather spend money on sexy hardware that won't solve the problem, instead of paying smart people who can design-in the uptime with combos of hardware, software, and procedures.

Quench-rant (for now).

Re:stinking devices

[Anonymous Coward]

[Blaming Management]

You want to place the blame on "Bosses" or "PHB's" implying that their ignorance is responsible for the problems that IT departments face.

The problem as I see it is one of accountability. A "PHB" can make a bad decision and that decision stands even in the face of disaster (or merely elevated costs). The person responsible is not being held accountable.

Reason #1: The job market isn't as soft as we've been led to believe, or else these places would have no staff after the first such decision.

Reason #2: There is such a huge division between the power given to "management" and anyone else in an organization (above OR below), that they are never held accountable for their actions.

3 seconds downtime??

[SeanAhern]

Someone with some embedded experience will have to explain this to me.

99.99999 percent downtime in a year works out to just over 3 seconds of downtime. Is that even possible? Feasible?

And if the machine is only down for 3 seconds in a year, who cares how long it takes to boot? Even .8 seconds to get to LILO will be too slow for it.

Re:3 seconds downtime??

[Xenu]

I think it is possible, but it isn't going to look anything like a PC. I'm used to seeing high-reliability systems specified as having 99.999% (5 9s) availability. You can tack on more 9s, but you are going to have a hard time proving that the system is actually that reliable. There are straightforward techniques for estimating hardware reliability. Estimating software reliability is much uglier. Statistical methods can be used to create system test plans and to interpret the failure data. The cost of the testing increases very quickly as the desired reliability increases.

Now if only....

[Picass0]

....if only I could get my X-windows profile to load in .8 seconds. Ahhh...

quick vs. accessable

[Anonymous Coward]

you know... I'd actually pay for an extended bios
in which you could designate a ethernet and serial
port to which you had access during the booting
process (a la Sun's RSC port).

Quick doesn't mean anything if the box is not
accessable while it's stuck in some sort of reset
boot reset loop.

You'd think the guys who develope BIOS would
actually be in tune with what is needed for tricky
remote support issues, but instead they're worried
about time to boot (think penis size vs more nerve
ending per square inch).

Re:.8 sec... SO?

[elefantstn]

Check it out along with CodeRedII attack info realtime

Dude, have you looked at your graph recently? You have some negative numbers at the beginning of August. Does your attack counter program subtract from the count when your server tries to attack other servers?

Re:Do the math once more. (DOH!)

[suso]

Nevermind, I forgot to shift the decimal place. Need some more caffeine. My mistake wouldn't have cropped up if we could trust slashdot authors on a more frequent basis. I just thought they had screwed up again.

Boot in 5 seconds

[Morris Schneiderman]

It's been a long time since I've used it, but my Osborne 1 used to boot from a floppy in about 5 seconds. Of course, there was no network, no raid, no scsi, no hard drive and only 64KB of memory to check.

Re:Boot in 5 seconds

[GiorgioG]

When I said "decent" OS I meant a feature complete OS - in today's world. ;-)

Fast Boot is also a User Interface issue

[drivers]

Even though this BIOS was intended for embedded machines, fast boot is also important for desktop PCs. Consider the Canon Cat designed by Jef Raskin (see "The Humane Interface" by Raskin). It takes a very short amount of time to boot, all you have to do is start typing and the computer powers on and loads the operating system, putting the cursor in the document exactly where you left it off. Not only that, but there is a hardware buffer for the keyboard so that it doesn't even lose the keys you typed while it was booting up. Now that is a computer designed with the user in mind. I'd like to make a PC operating environment and the first thing I'd do is make sure it boots fast. I was thinking the BIOS would be the slow part but if it's possible to speed that up, then that is all the better.

Re:Fast Boot is also a User Interface issue

[ethereal]

Yeah, but who wants to type when you can't even see where you were yet? The buffered keyboard is overkill; no normal user is going to start typing before the display comes on and the last known document cursor position is displayed.

Re:Fast Boot is also a User Interface issue

[drivers]

Whatever you type is the "current selection". When you're done, just cut and paste to where you want it.

BIOS with Serial port console?

[garver]

On the BIOS topic, do you know where I can find BIOS that talks via ttyS0 instead of the video card?

To me, this is a requirement for servers. It means I can completely administer it remotely via a cheap terminal server. It also means I can dump the video card in my servers. Sure, Linux can use the serial port for its console, but that doesn't help me when Linux isn't booting. I know there are remote KVMs, but they aren't the cheapest thing in the world.

Re:BIOS with Serial port console?

[Anonymous Coward]

http://www.realweasel.com/ Serial Console for your PC..

Re:BIOS with Serial port console?

[geirt]

General Software ..... :-)

Yes, embedded systems need 0.8 sec boot time *and* console on /dev/ttyS0, and the General Software BIOS supports both.

Re:BIOS with Serial port console?

[The Madpostal Worker]

I bought a box from VA Linux, and its bios had an option to redirect the console to /dev/ttyS0 (and the EMP port on S1)

Win 9x and re-booting...

[Robber Baron]

Great! Now I won't have to waste as much time sitting through all those painfully slow Windows re-boots!

Re:Great for embedded devices, but not for other..

[Cock Knocker]

The microwave was supposed to be a box that you put anything uncooked into, you push a button, and you get cooked food out of. The fact that you go to your stove or grill, and the taste of leftovers is evidence that it has failed. I don't care if the new system is embedded in the box, or offloads the work to a bunch of these generic computing devices, but intelligent devices are the future.

I'll argue the scanning fridge point too - I predict in 15 years a sizeable number of people will pick a constant stock of food that will be maintained by automatic shipments, and people will only go to the store for emergency (ran out of milk unexpectedly) and custom (feel like trying this new kung pao chicken recipe tonight) stuff; both of which the system updates its rules in response to. We have the technology to do this now, but people obviously need to get used to the idea. Why exactly do you consider this scenario to be absurd? What /I/ consider to be a waste of time is going to the grocery store every time I need food.

Automation and other use of technology to overcome what many people consider to be inconveniences is not overkill. Robustness is desireable, but not always at the cost of eliminating tremendously useful feature sets. Embedded-QOS is nearly achievable by a single modern desktop if you lock it down & it's dedicated, and maybe with a couple of them for redundancy. Either way I'd sacrafice a few second reboot once a year in order to be able to come home after work every day to a full fridge, pull anything, put it in the Universal Cooker, push a single button and have any possible meal cooked exactly to taste at the microscopic level.

ka-pow!

check it out

[brad3378]

Not quite as fast, but still very cool
http://www.acl.lanl.gov/linuxbios/

The Dual Athelon setup sounds promising

Re:check it out

[norton_I]

Actually, both this solution and LinuxBIOS can get to a command prompt (or, more likely, load an embedded application) in 3 seconds.

The 0.8s to LILO prompt is nice because it demonstrates how fast the BIOS is, but is irrelevant from a system designer standpoint: The important thing is how fast they can get their program running.

Doesn't really matter...

[Penguinoflight]

The PII and PIII are identical in architecture (until coppermine), so, who cares if it was a PII or a PIII, for that matter, who cares if it's Intel?

Re:.8 sec... SO?

[stikves]

Yeah, it's definately faster than 0.8 secs. because you start booting linux in 0 secs. (Even no lilo).

But i head that it required you to reboot the machine if it booted "before" HDDs were up! (I do not know the current situation. I haven't checked it for more than a year).

Re:.8 sec... SO?

[mattdm]

You can *easily* fit all drivers to boot almost any given system in 8MB. In fact, you can almost certainly do it in one meg. What you can't do is have all the drivers in existence availabe -- but who needs that?

Got Math?

[Kasreyn]

Let's get out our calculators, class...

365.24 days per year (from Space.com. I don't know of any more accurate (more decimal places) numbers than this. Even if you were to add 5 whole days to the year though, it won't even add one one-hundredth of a second to the final result, so I think we can go with this).

99.9999999% of 365.24 = 365.23999963476

365.24 - 365.23999963476 = 0.00000036524 days

This is the maximum allowed downtime.

Assuming a day is exactly 24 hours long (I'm fairly sure it is),

0.00000036524 days = 0.00000876576 hours

0.00000876576 hours = 0.0005259456 minutes

0.0005259456 minutes = 0.031556736 seconds.

Thus, 99.9999999% percent uptime requires NO MORE than ~0.0315, that's three hundredTHs of one second, downtime, per year.

Nope, don't think we're there yet, but you keep pushing that 99.9999999 number if it makes you look good. After all, the general public can't do math either... =)

-Kasreyn

Re:Got Math?

[pkesel]

At least read posting, if not the article. It's 7 nines, not 9.

Re:Got Math?

[DarkMan]

Your close.

Your calculating for 9 9's. They count the two before the decimal place too.

If you allow for 100x as much downtime, you;d get 3.15 seconds, for 7 9's. Which is what they quote.

Re:Got Math?

[Anonymous Coward]

After all, the general public can't do math either
Present company included

Re:Got Math?

[8-Bit Junky]

Your calculation involved nine 9s, not seven. Got Math, indeed.

Re:Got Math?

[Benjamin Shniper]

That's nine nines (99.9999999%)

seven nines is (99.99999) and that makes your calculations 1/100th of what they should be. The final answer should be about 3 seconds/year in uptime.

Uptime of the computer, I suppose, is what this handles, although I don't believe the OS boots in .8 seconds, just the bios. And the server you may want to keep up might not boot in the .8 seconds either.

Plus this doesn't deal with other problems such as why the computer went down in the first place.

-Ben

Re:Got Math?

[dbretton]

Let's put our calculators away, and get out our
fingers and toes...

let's see: you calculated for 99.9999999 % uptime.

That would be:

9 9 .9 9 9 9 9 9 9

^ ^ ^ ^ ^ ^ ^ ^ ^

1 2 3 4 5 6 7 8 9

now let's get out those calculators:

9-7 = 2

wow! 9 - 7 != 0!

You just calculated uptime for "9" 9's.
Apparently your math problem is much more severe than that of the general public.

-D

Re:Got Math?

[Anonymous Coward]

That's arithmetic, not math. If you got math, you'd be able to figure out 365.24 on your own:

365 days per year

Plus 1 day every 4 = 1/4 = 0.25

Minus 1 day every 100 = 1/100 = 0.01

Plus 1 day every 400 = 1/400 = 0.0025

Total = 365.2425 days/year

Re:Got Math?

[MrScience]

that makes your calculations 1/100th of what they should be. The final answer should be about 3 seconds/year in uptime.

Hmmm....

Re:Got Math?

[Anonymous Coward]

> Even if you were to add 5 whole days to the year though, it won't even add one one-hundredth of a second to the final result.

Sure it will.

Re:Got Math?

[Galahad2]

The leap year system isn't accurate. You don't think the Earth would've been so cooperative as to be only 1/365th of a rotation off? A year is in fact 365.242198781 days (365 days, 5 hours, 48 minutes, and 45.9747 seconds). So much for math. Maybe physics?

Re:Got Math?

[Galahad2]

Er, my mistake. I meant to say: "You don't think that Earth would've been so cooperative as to be only 1/1460th of a rotation off?"

The leap year system is a day every four years, excepting multiples of 100, excpting multiples of 400 (but I didn't want to do the math to find out that fraction).

Re:.8 sec... SO?

[Telek]

funny =)

Umm, I just ran it and it reported fine, no negative numbers.

Maybe it can't hold more than one person accessing it at once =). It wasn't exactly written to be robust and hold load. I dunno, I figured that it wouldn't be around for more than a few days so I didn't exactly code perfect...

not only crashes

[Anonymous Coward]

actually, this would be great for casual "embedded" uses as well, such as a computer being used in a car for GPS navigation. You want it to come on with the car, and not have a wait for a boot. Same with a computer being used as an MP3 player in a home stereo system. You want it to act like a component. Hit the power button and it is on, like your CD player.

.7 seconds to hit F2 or DEL

[simetra]

It's nice having a few seconds to hit F2 or DEL to go into the BIOS if you need to change something.
This is obviously for the Suits.

0.8 seconds! wow!

[Kragg]

That means, if you add the 75 seconds for windows to boot, you get... oh. about same as it was.

Re:.8 sec... SO?

[elefantstn]

Here, I took a screenshot of it: here. I think it might just render different on Mozilla.

Guess You Don't Ever Patch?

[Anonymous Coward]

So, you have machines that have been up for over a year without ever being rebooted.

Uh.. so I guess you never apply any security patches or bug fixes eh?

Glad you don't work for me.

Re:Guess You Don't Ever Patch?

[Wakko Warner]

I'm glad I don't either. You seem to be super-paranoid and prone to wasting time with needless security issues.

When in my previous post did I say these machines were Internet-accessible, or even mission-critical?

- A.P.

MRBIOS

[bored]

News, hmmp, check out MRBIOS. I first discovered them back in 92/93. Back then they had auto IDE detection, support for big IDE drives, and of course a FAST boot option. A year or so after that they had (software like the promise) IDE RAID support in the BIOS. Today I still have a VLB 486 machine (my firewall/webserver) with MRBIOS. It has a 60 Gig harddrive and a 16.7 Gig harddrive plugged into it with a massive uptime ratio (greater than a year and a half at up at one point). The machine is sub .8 second warm reset times. Its basically instant. The screen clears and lilo starts booting linux (I have lilo configured not to stop unless the shift key is held down) if I press the reset button. If the machine is comming up from a cold start the bios flashes post for something less than a second and then displays a flashing "Waiting for Harddrive to spin up" while the harddrives are going Whhhhhhhhmmmmmmm... As soon as they sound spun up the machine starts booting. I have machines that are 3 years old that don't support 16 gig drives and this little box is getting on towards 10 years old and it has a 60 gig HD plugged into it. I put a dx4-100 overdrive in it a few months back and the board which was bought right when the Dx2-50's first appeared. Poped up and said,"Newer than Dx2 486 at 99mhz" ,or something like that.

Its really sad though that these guys never caught on. Most of the 'cool' bios features that have appeared in the last few years in award/AMI were in MRBIOS in the early 90's. Now they are just a shell of a compay and they don't have BIOS's for machines newer than a few years old. Some people are just ahead of their time, Well I guess i'm going to go home and reboot my machine from the third harddrive now... lol..

Re:MRBIOS

[coli]

"Its really sad though that these guys never caught on"

That's capitalism for you. The best product don't nessasary wins.

They pointed it out: "PIII" + "400" = "underclock"

[yerricde]

If it is an underclocked chip they hould point out that fact.

They probably figured that it should be obvious that any chip labeled "pentium !!!" running at 400 MHz is running less than its stated clock speed.

The first runs of PIII added only a larger cache and the new SSE vector instructions.

Re:.8 sec... SO?

[LMariachi]

elefantstn was probably interpreting the red numbers outside the short bars as negative, since that's how they're often displayed on an accounting spreadsheet. The color of the short bars themselves doesn't help, either: color differences should be used to convey information.

Take a look at the Edward Tufte books, particularly The Visual Display of Quantitative Information.

2.4.8 on 08/11/01, 2.4.9 on 08/16/01

[TeknoHog]

If Linux is so darned stable, why do you guys care?

You mean the guys who upgrade the kernel every week?

General Software sucks

[Eric Seppanen]

I used to use General Software Embedded BIOS. It was truly awful. It came with source code (note it's NOT open source), and there were nasty bugs in the code, in files that hadn't been touched in years. The performance was awful, the development kit was clunky, the documentation was poor, and the code was crap. My impression was that of a company that spent very little time developing software, and were mostly interested in squeezing the last few dollars out of an old, mostly unmaintained lump of code that had very little value left in it.

The idea that someone today would actually _want_ a 16-bit legacy BIOS in a newly designed embedded device is laughable. The only reason to do it is if you want to run embedded DOS (gag!) or Windows, which ain't likely if you're shooting for 99.99999 percent uptime.

Now I use a variation of Linuxbios. It works great, it's free, and I'm free from debugging somebody else's crappy assembler.

LinuxBIOS at least as fast

[sjames]

LinuxBIOS is at least that fast and is GPL. It's also been around for at least a year now.

Re:LinuxBIOS at least as fast

[sjames]

Perhaps this link will work better.

I don't think there is such a thing

[Skuld-Chan]

as a PIII 400 - as I recall the first P3 was the 450.

Re:.8 sec... SO?

[flatrock]

This BOIS is a building block to be used in a system that needs to power-up quickly. An example of a place it might be used is a set-top box. People don't want to wait 20 seconds for their TV to come on, and people have also become a lot more aware of appliances that don't really shut down when you turn them off. I can see a market for this BIOS, but I don't really see it in the high availability market.

In my limited experience with a telecom product that needed 5 9s uptime, everything had a level of redundancy, because you had to assume some hardware was going to fail. That means that you are possibly running at reduced performance while the system is comming back online, but the system doesn't go down just because part of it was rebooted. The 79s thing sounds like something someone in marketing though sounded good, even if it's not that applicable. It has that cool buzzword, marketing feel to it.

Only a linux twit...

[slonob]

would find this interesting. it's not the BIOS that is cramping your style, it's LILO.

I use Linux and I use FreeBSD. The FreeBSD boot loader rules. And, err, it's not the lack of journaling that makes ext2 suck. UFS is not journaled and yet does not ask you to wait as long as ext2 on a sloppy reboot.

Well, crap

[Kasreyn]

I thought they meant 7 nines after the decimal place. =P

Hehe.

-Kasreyn

UltraSparc is slower clock for clock anyway

[throx]

Who was comparing clock for clock? I was comparing SPEC benchmarks.

Oh, and you are wrong about clock-for-clock too.

500MHz UltraSPARC-IIe (Blade 100) SPECInt 165.
1400MHz Athlon SPECInt 495

The UltraSPARC is SLOWER clock for clock than the Athlon!!!

Re:UltraSparc is slower clock for clock anyway

[Anonymous Coward]

Why are you comparing integer performance?

Re:UltraSparc is slower clock for clock anyway

[throx]

Seemed appropriate, especially if you plan to be using the system for development/compiling (a pure integer application). Were you really planning on getting a Blade 100 for floating point applications?

If you want to compare FP performance,

P4/1700 is 598
Athlon/1400 is 426
Blade 100 (500Mhz) is 163

So the UltraSparc is marginally faster clock for clock on FP performance than the Athlon but still gets whipped by the P4. Oh, and just before you accuse me of changing CPUs, the P4 beats the UltraSparc IIe clock for clock in integer performance also, just not by quite as much than the Athlon.

Re:UltraSparc is slower clock for clock anyway

[Anonymous Coward]

Now compare bus performance. Ouch. Intel gets hammered. And what's the biggest bottleneck in modern computing? Not integer speed, mate.

Re:UltraSparc is slower clock for clock anyway

[throx]

This is just getting funny now!

The Blade 100 has a 100MHz 64 bit bus (from CPU to memory) - exactly the same as a Celeron. A Pentium 3 has a 133MHz 64 bit bus, the Duron has a 200MHz 64 bit bus, the Athlon has a 266MHz 64 bit bus and the P4 a 400MHz 64 bit bus.

So, the Blade is the equal to the Celeron and gets whipped in bus performance by every other x86 architecture around.

What else shall we compare?

Re:UltraSparc is slower clock for clock anyway

[orangesquid]

SGI's have a very fast bus. My MIPS-4600/200Mhz box is whooped by my Celeron/633Mhz -- except when it comes to any sort of memory transfer.

The SGI bus can handle 2.2G/s.

Let's see an intel box pull *that* one off :)

Re:UltraSparc is slower clock for clock anyway

[throx]

You mean like the P4's 3.2G/s bus?

Also, if I recall we were talking about the Sun Blade that some AC said was a much better system than any intel box...

Re:UltraSparc is slower clock for clock anyway

[Anonymous Coward]

Ok, all this tech talk lost me, just summarise for folks and tell who has the biggest dick.

Re:UltraSparc is slower clock for clock anyway

[throx]

To summarize, I think you'll find the x86 has the "biggest dick" (at least as a result of these comparisons), but it's pretty goddam ugly and no woman who wasn't paid would go near it.

Re:UltraSparc is slower clock for clock anyway

[orangesquid]

But the P4 hasn't been around since 1994, like my SGI machine has. No doubt SGI's have faster busses now, hah.

In fact, I'll link you: System bandwidth: 716GB/sec
Mwahahaha.

Re:UltraSparc is slower clock for clock anyway

[throx]

Hell yeah.

But saying an Onyx3800 is faster than a P4 isn't exactly the same as saying that we shouldn't be buying x86 machines because a Sun Blade 100 is $995.

Personally I can't understand why ANYONE would want a Sun Blade 100. It's underpowered, overpriced, old technology and doesn't really fill any market you'd need a Sun Workstation in. The original poster's comment that SPARC (in this case an UltraSPARC IIe) kicks anything in the x86 world was just false any way you looked at the number.

An Onyx3800 on the other hand has some very good reasons that people would want to buy it. Shame it's a little more expensive than $995.

Dead Nietzsche asks:

[Anonymous Coward]

What is the use of embedded linux in a microwave ?

Only morons will answer "cat".

Wrong

[Anonymous Coward]

Unless seven nines is 99.99999 and not 99.9999999 which you stated.

What abour MR BIOS?

[SaDan]

I used to have some old 486 motherboards that had MR BIOS, and they could boot in about a second or so. You normally had to wait for the hard drive to spin up when cold booting.

But I do find it sort of funny how as you get into machines with loads of processor power, it takes them eons to boot. Sure, self tests and all that stuff are nice to have, but it's still funny.

Re:.8 sec... SO?

[Anonymous Coward]

Read the article (or summary) - it said EMBEDDED, not desktop, systems.

6 seconds to missile impact is a tight CPU budget

[jkeene]

Embedded systems often need good recycle times. A weapons programmer gave me the title quote, wish I could attribute it but it really made the point to me.

Re:.8 sec... SO?

[aozilla]

You can definately do it in 16 megs since I've seen a diskless linux box run on just 16 megs of ram. What I'd love to see is a barebones vmware-like OS done in ROM. I'm using windows on vmware on linux now (in raw disk mode) and I forget sometimes that I'm not windows directly, it's that fast (linux on windows sucks however). Of course my stupid BIOS doesn't let me stop the memory check so it takes me god-knows-how-long to check my half gig of memory (vmware loves memory).

With a vmware-like device you could easily store a memory dump to disk after bootup, and then load that directly into memory every time you restart (unless you need to update drivers).

Re:.8 sec... SO?

[blair1q]

Desktop operating systems are designed for configurability, not zippy boots for HA. It has reams of code that's only there to make reconfiguration easier. You can pre-set a desktop kernel with its fixed peripheral configurations, but there are embedded OSes designed for that.

Linux would only be repurposed to do this sort of thing because it's open source, and because hackers understand its operation. That doesn't make it the right tool for the job.

--Blair

No Kidding

[CMiYC]

I think is goal was to simply give an extreme example. No kidding you don't want this type of embedded system in a brake control system.

Re:No Kidding

[Telek]

OK, fair enough, but seriously then, what are some specific real-world applications of such a device?

Re:No Kidding

[spudnic]

Linux based car mp3 players come to mind immediately. Who wants to wait 10 seconds after they hit the power on button before something happens.

Re:No Kidding

[Telek]

This may be a crazy question, but how hard would it be to write an MP3 player into the bios? Or even just something that do simple bios stuff, then load the MP3 player code off the start of the hard disk or CD, then just have a simple MP3 player interface (for example, with a numeric keypad to change songs)?

Re:No Kidding

[rew]

what are some specific real-world applications of such a device?

There are lots of devices that you expect to be ready for use within a second or so after you turn them on. More and more of these devices are getting more and more complicated. To the level that these would run a real operating system like Linux.

I've tested a "digital TV decoder" box. It would take at least 20 seconds to boot. So you turn on the box, and have to wait quite a while before it can be operated. It was slow as hell in anything you'd do, and the guys writing the software hadn't heard about "user interface design". So for instance, if you pressed "channel up" it would take a couple of seconds for the device to react. If that's a technical restriction on digital TV channel decoding, then that's fine with me. But in the meanwhile you really NEED to provide feedback about the channel switch by updating the front panel!

Roger.

Re:.8 sec... SO?

[Telek]

aah ok, the problem isn't negative numbers, the problem is that you browser doesn't render DIVs correctly... I use a div for the graph with a height setting corrosponding to how high the graph should be (concept! :>) .. I had this problem on mozilla to, it has a minimum height apparently, and if you specify less than that height it displays the box much bigger than necessary. Try to load it with IE or netscape (dunno about the latter, but I know that it works in IE). Thanks for the feedback thou =)

Pause for effect.

[dozing]

Time to boot bios 0.8 seconds.
Time to boot linux 3.0 seconds.
Finishing your work before the windows user in the next cubicle finishes booting: priceless.

Re:Pause for effect.

[Anonymous Coward]

Shut up, fanboy.

Re:.8 sec... SO?

[csbruce]

My VIC-20 will also boot up in .8 seconds, and it only has a 1-MHz processor, and it'll go all the way up into the command-input mode!

READY.

Re:.8 sec... SO?

[Ioldanach]

you think that you can fit everything that you need for a linux or windows bootup in 8MB of flash? You think that you can fit BOTH of them?

floppyfw (Firewall on a Floppy) runs a linux 2.4.5 kernel. It handles ipchains, iproute2, traffic shaping, and can retrieve ip's for all its nic's with dhcp. It's held on a 1.44 MB floppy, so yes, I think I can fit everything I need for a linux bootup in 8MB flash. I might not be able to store *all* the modules I need, but I can certainly store enough to start the machine with.

Re:.8 sec... SO?

[Telek]

with memory being DIRT FSCKING CHEAP... half gig isn't that out-of-this-world anymore, eh?? =)

Umm, yes, a totally stripped down and trimmed version of linux could run in 16MB, but a fully fledged and usable version? I doubt if it could fit in 16MB. Do a memory dump sometime to see how much memory your system is using right after bootup, I'd be interested to see actually.

You can't just do that load-a-memory-image trick because all of the devices in the system need to be in the exact same state that the drivers think that they are in. Thus if you were to just load a memory image, all of the drivers in the system must support hot-reset, including kernel drivers. Windows 2000 finally supports that (hibernate mode), so you could hypothetically boot up, immediately hibernate, and then just always use that hibernate restore. that would speed things up considerably. My system uses about 70MB total on boot (but I'm sure that if they were smart and didn't load stuff that "might be used soon" it could be 40MB. Compressed that's easily 30MB which is just a second or 2 to load off the disk. That's gotta be faster than doing a full boot all the time.

Anyways, it's not just as easy as it sounds. I don't think linux has it yet, does it? Is it in the works? Do linux users even want it?

I don't know why they don't implement that as a standard feature, but I'm sure they have their reasons.

I guess that's good for embedded systems...

[Anonymous Coward]

However, from hitting power until the time I have complete control on my win98 box is ~22 seconds. I relish those seconds as my personal warmup time. Booting in 5 seconds would be kind of... jarring.

Re:99.99999%? That's great...

[Anonymous Coward]

so did the HPUX reboot ?

Re:.8 sec... SO?

[Telek]

OK, I stand corrected!! =)

But seriously then, what's the speed difference between that and just loading the image off the disk?

Booting a mainframe takes HOURS.

[CrystalFalcon]

This society is fucking amazing sometimes. :-)

One half of this discussion concerns how reliable mainframes are. The other half concerns how quick reliable systems boot, must boot, or should boot. People seem to universally agree that the two are holistically connected.

Well, guess what: a mainframe takes HOURS to boot. It runs forever, but it also takes forever to start. I used to run a military-grade telco switch that took about three hours from powerup to in-service. How is that for a boot time?

Sigh, I miss my Commodore 64. Flick the switch and you were set to go. :-)

Wow

[ioman1]

So when will this BIOS be available to consumers?

Re:.8 sec... SO?

[xwred1]

What?!

8 entire tenths of a second?

But I want it now!

Re:.8 sec... SO?

[xwred1]

I read something about the Linux kernel summit a while back, saying that hibernation support was on the plate for v2.5.

BeOS

[Jaysyn]

...and I thought my 15 second BeOS boot was impressive....

Jaysyn

Re:BeOS

[vortexau]

Hey - that's MUCH quicker than my Amiga OS3.1 at
20 Sec (with no CD in Drive), and most certainly,
much more than the 40 Sec of OS3.5!

But, of course, 3.5 is a Patched Double-Boot! And
the hardware is pure 1987 (with 1991/2 CPU)!

Roll-on A1 with G4 or G5!

Regards,
JK

99.99999 is five 9's, not seven

[Anonymous Coward]

Seven 9's would be 99.9999999

you don't count the first two. M$ lies. They call 99.999 five 9's, that is not true and I doubt they can ever get that on an actice system

Re:.8 sec... SO?

[Anonymous Coward]

> you think that you can fit everything that you need for a linux or windows bootup in 8MB of flash?

Yes. Stripped down kernel + a couple of modules. Compressed. Waaay below 8Mb. Waay below 4 Mb. Way below 2Mb. I can fit that in a 1.44 floppy.

Of course, it buys absolutely nothing, as, to get a usable system you need daemons that are on disk, so you'll have to wait until the drive starts spinning. You also have to scan for hardware, but there have been a project to get this down (ie: don't scan for hardware, remember previous hardware found and just probe it).

Sure, it is possible to put userland tools in those 8Mb flash too (people tend to forget that 8Mb is a lot of space), but as a MP3 player is not usable before it can access the MP3 files, so it doesn't buy much. In general, most interesting applications are not usable until the disk spins.

Having all the userland stuff stored continuously on disk on a proper location (as every area of the disks don't have the same transfer time) could certainly be a plus (to avoid needless seeks at boot time, you can 'preheat' that disk content by reading it in a single pass: it will end in the memory cache and you'll get zero ms seek time). Of course, this root device should have no space left, and only contains things that have 100% probability of beeing accessed at boot)

* Spin disk at power on
* High perf BIOS
* Kernel compressed in flash, copied and decompressed in memory
* Booting kernel only probing expected devices (everytime the devices changes, you have to re-flash the bios). This may buy you nothing, as you'll have to wait for the disk at the next step.
* Wait for disk to be ready
* Pre-heat the root device (At 20Mb/s)
* Start userland boot

Cheers,

--fred

Re:.8 sec... SO?

[xanadu-xtroot.com]

Ah... those were the days. I remember my (second :-) ) C=64. I tried using Geos. 2 5.25" floppys only to gain not much when I could go across the street to a friend's house who had one of the first Macs... :-)

Tandy 1000HX

[fishbowl]

Am I the only one here who remembers DOS3.3 machines that had DOS in ROM? By the time you took your finger off the power button, you were at the prompt, or your autoexec had run.

Re:Tandy 1000HX

[adolf]

I was reminded of the old PCs just last night, when trying to configure LILO on a new PII-based Linux box.

On the first reboot, it came up, POSTed, and acted normal. Instead of LILO, what instantaneously appeared on the CRT was a message saying "CANNOT FIND ROM BASIC - SYSTEM HALTED".

It was beautiful, in a 40-column white-on-black font that I hadn't seen used in years.

I stared at it in awe for several minutes before rebooting to see it again. =)

Re:.8 sec... SO?

[Telek]

yeah, Windows XP monitors file access on bootup and rearranges files loaded on bootup in the same order that they're loaded on the hard drive, so after a few boots your system actually boots quite a bit faster, it's impressive. One of those "Duh, why didn't we think of that before??" =)

Compared to your boot process there, what's the difference between that and just a

* Spin disk at power on
* load simple BIOS that loads and uncompresses first xMB off the HD into ram
* execute from extracted.

I don't think that'd be much slower (if any slower at all, since you have to do the disk read anyways), but this way you don't have to worry about the whole BIOS thing.

Here's another silly question. Can't you postpone most of the BIOS checks to happen after the kernel has started loading? I.e. once you've decompressed the root kernel image from the HD into memory and the base kernel is loaded, do some of the less-important BIOS checks concurrently then (i.e. upper range memory check, floppy seek, etc, etc).

Re:.8 sec... SO?

[fishbowl]

> with memory being DIRT FSCKING CHEAP... half
>gig isn't that out-of-this-world anymore, eh??

NVRam is not "dirt fscking" or any other kind of cheap. Also, RAM for embedded devices is not always cheap like it is for consumer devices.

Since we're talking about embedded devices, I think it's fair to point this out.

>Umm, yes, a totally stripped down and trimmed
>version of linux could run in 16MB, but a fully
>fledged and usable version?

It depends on what you mean by "usable"

I have a notebook I use all the time with only 16MB. It's nothing like "totally" stripped down.
Trimmed, as you'd do for any notebook install, but not as much as you seem to think. The same notebook runs windows95 just fine too.

Re:.8 sec... SO?

[Snodgrass]

There's a guy here at work (unnamed) that has successfully embedded a kernel in the bios of a Tyan 2390 motherboard. He says that he can type 'init 6' and have the login prompt back up in less than 2 seconds. We build cluster systems, and the reason for the project is so that with Linux controlling everything, it can report exactly why a node failed during boot up...even if it never POSTs!

I'm no techie...I just build the things, so I don't know much about all the technical stuff...but it's pretty amazing!

Re:Get a Mac?

[doorbot.com]

The major problem is the chipsets and the 20-year-old designs they're based around. Drop in full 32-or-more-bit DMA controllers or require all peripherals be bus-master capable, segregate the ISA bus to its own out-of-the-way 16MB window somewhere (see: Apollo DNx000 family), hardwire a handful of interrupts and a hardcoded address range to each slot (see: EISA), drop the legacy keyboard/mouse interface, and redo IDE entirely (see: SCSI). While we're at it, let's scrap BIOS and replace it with OpenBoot. Now there's a machine free of legacy crap that might be worth writing home about.

Geewiz, that sounds like my Mac (G3).

But you're right, there is a need for a chipset redesign, not a processor redesign (at least for now). Thankfully, Intel has the Itanium. It's a start, at the least.

My question is, what do IBM's Power-based systems use to boot?

Re:6 seconds to missile impact is a tight CPU budg

[Anonymous Coward]

Here's a real world example for some work we are doing on the new F/18 Super Hornets. The Navy came to us with the desire to add an extra redudancy to the system that controls the HUD. Currently it is a single redudant system, and if both systems go offline it takes aproximately 10-12 seconds for a full cycle and the system to come online again. Naturally, in combat 10 seconds is an eternity to be flying around with no HUD, so we were recently given the task of adding the ability for our fuel control box to also drive the HUD until the other systems come on line. So, quick boots are extremely important.

Re:.8 sec... SO?

[Telek]

NVRam is not "dirt fscking" or any other kind of cheap. Also, RAM for embedded devices is not always cheap like it is for consumer devices.

The comment I was replying to was that the person had a half gig in their home machine, and since 128MB PC133 is $18.95 at the corner store now, fully populating your motherboard isn't expensive AT ALL. I can't believe how fast and quickly memory prices have fallen!! Can anyone else remember when it was like $30/MB back with SIMMs? Then with the DIMM price hike? Damn... So yes, SDRAM is dirt cheap right now =) Kinda pisses me off. About a year ago I paid $400 for 256MB RDRAM off ebay for my P4. Now it's like $81 for the same amount.


I have a notebook I use all the time with only 16MB. It's nothing like "totally" stripped down.
Trimmed, as you'd do for any notebook install, but not as much as you seem to think. The same notebook runs windows95 just fine too.


And what do you run on it? No, seriously. I want to know what you can run on a 16MB laptop, deemed "usable". If you're referring to maybe wordpad or 1 small old application, ok sure. But I remember with 32MB on a laptop trying to run Word with Win95 and it choked bigtime. Took forever to load, was ok to use provided that I didn't try to use any other application at the same time.

I think the point of this entire thread was that they were talking about doing this to your normal home machine to speed up booting time. I'm just not certain that it wouldn't be a lot easier, and no slower, to just put that block at the beginning of the hard drive and load it from a very simple bios. This way you're unlimited as to what you can do, you don't have to flash to test anything, and since you have to read stuff off disk anyways sooner-or-later during the bootprocess, do it all at once at the beginning and grab your bootstrap/ramimage at the same time.

Re:99.99999%? That's great...

[spudnic]

4 year uptimes for NetWare are not that uncommon, especially in a controlled environment where they are assured of consistant power.

Re:99.99999%? That's great...

[Anonymous Coward]

liar! How could you have seen a 700 day uptime? Windows 2000 hasn't been out for that long yet!

I'm a 5-0 user

[pacc]

Since my linux system only is switched
on when I read mail I'd rather have a more
robust handling of unexpected shutdowns.
(e.g me pushing the powerbutton)

Whenever I do this all kinds of apps leaves
.pid files i my /var directory and until they
are deleted the apps refuses to start (e.g vnc)

Computers without a shutdown-menu rules!
All Hail ext3-fs which won't stall for 15 minutes
at startup.

Re:.8 sec... SO?

[aozilla]

Umm, yes, a totally stripped down and trimmed version of linux could run in 16MB, but a fully fledged and usable version? I doubt if it could fit in 16MB.

The idea is to put a minimal kernel (no apps) in the 16MB, and then DLKM the rest. Actually, my idea was to put a minimal linux with just enough support to run freemware or some other vmware-like product, directly in the kernel. I have no idea how much space that would take, though. It's basically a microkernel architecture, the meat of the OS (filesystems, networking support, etc) would go on top of it. Besides booting quickly, you'd hardly ever need to reboot in the first place, because the majority of the code will be dynamically loadable.

You can't just do that load-a-memory-image trick because all of the devices in the system need to be in the exact same state that the drivers think that they are in.

I did this all the time with vmware. I suspend the linux to disk, reboot windows, then resume. The TCP gets a little messed up, but it's quite simple to just bring the interface down and then up again. Pending disk I/O is presumably flushed before halting the OS execution, I'm not sure exactly how they do it but they do, and I'd imagine most of their work is spent making it compatible with guest operating systems, if you had the actual support of the guest operating system it'd be a lot easier.

The only potential problem I see for this approach is with gaming. Perhaps it could be worked around with raw I/O access and direct screen writes, or maybe you'd have to modify the intel architecture itself, I'm not sure. But other than gaming, introducing a small amount of latency into system calls is worth it IMHO for the gain in reliability. Again, I'm running windows on vmware on linux and I am noticing zero problems. This on a 500Mhz celeron with half a gig of memory (256 megs dedicated to windows). An earlier article mentioned about how people buy machines which are way too powerful to be used, I think we've reached the point where we can start trading processing speed for features.

Re:99.99999%? That's great...

[Breakfast Pants]

You are indeed correct, you probaly read about it in The Hacker Crack Down.

Forget the math

[mattyohe]

Would you really want a SOYO board for a mission critical operation?

Re:.8 sec... SO?

[Anonymous Coward]

and the silly fool is still running ext2

WTF? Last year the Linux advocates would have us believe that ext2fs was comparable to NTFS. Now it's a piece of shit? What changed? [Other than the release of ReiserFS]

Re:Great for embedded devices, but not for other..

[ahde]

Bah! My microwave is 26 years old!

Re: Never heard of the Linux BIOS project?

[Red Moose]

You can do this; of course, it wipes your original BIOS stuff, but you can put a kernel in the BIOS and have almost instananeous booting.

Google search to find the home page.

So I could upgrade my car MP3 player to PIII now?

[Powercntrl]

A few years ago (can't remember the exact year, but I could look up the receipt for the motherboard) I decided to jump into the car MP3 player craze... Since I frequently made round-the-corner trips, I didn't want something that would take forever to boot. I did a lot of talking with friends about custom-designing something, but ended up using off-the-shelf (actually, out-of-the-discount-pile) PC parts. Basically, I settled on an old Intel-made FX chipset socket 7 board with intergrated Vibra 16 sound, 16MB of RAM and a Pentium 90MHz downclocked to 75MHz to reduce heat.

For storage, since I had just gotten one of those cheapie $199 2x CD-RW drives, I decided to use cheap, reliable CD-R media and an old 4x CD-ROM drive. I've seen FAR too many hard drives die in much more comfortable enviorments than the trunk of a car, so a hard drive was out of the question. Since this motherboard was old, it didn't have a CD bootable BIOS, so I figured I'd boot from a floppy. After all, it only needed DOS 5.0, the ATAPI CD-ROM driver, MSCDEX and MPXPlay (a very versatile DOS MP3 player) - space wasn't an issue. It worked fine but took forever to boot.

I then remembered that there were flash drives for embedded computers that required no spin-up time and are resistant to heat and vibration. I bought a 8MB M-Systems IDE flash drive for about $90... I look back on that price and cringe. This improved the power-to-music time to about 30 seconds - better, but not perfect.

It then dawned on me that since the motherboard I was using was old, there may be an MR BIOS available for it. Back in the day, I ran a BBS and still had all the files archived... MR BIOS was once distributed as shareware. Sure enough, I had a MR BIOS image for that motherboard. Since the flash drive requires no spin-up time, the system boots INSTANTLY. I'd try and stopwatch it, but I don't think my reaction time is quick enough. The system was now able to go from power on to MP3 playback in under 7 seconds, a process that involves booting DOS, loading CD-ROM drivers, loading MPXplay and scanning the disc for MP3s. I figure if I found a faster loading CD-ROM driver or bought a faster drive, I could get this total start time even lower.

Nowadays this hardware is worthless, but back when I got it, it was just cheap. MPXPlay has been updated since I started using it and now my car MP3 player also supports OGG. It is nice to know that if a new format comes out that requires a LOT more processing power, I can make a PIII boot just as fast as my old underclockded P90. :)

The above is a joke

[WD]

"Insightful" .... sheesh!

Re:99.99999%? That's great...

[Jason Pollock]

You are right. However, while they are run in duplex mode (and simplexed for upgrades), it still involves a reboot of each individual board to do the upgrade. Reducing the amount of time spent in simplex is something that manufacturers spend a _lot_ of time (and money) doing.

Jason Pollock

Uh, dude, those files have been deleted from /code

[devphil]

Perhaps you missed, or chose to ignore, the fact that the links you posted to the CVSweb filter are calling files out of the Attic.

The Attic is for deleted files.

So, bitchslap and modslap aren't being used anymore. Yeah, the spirit of consorship probably lives on in other pieces of /. but you do your cause a disservice by pulling up deleted code and treating it like it's live.

Linux Bios has been there, done that

[ColaMan]

Who needs LILO?

Check out the linuxbios home page

Cold Boot to single-user in under 3 seconds.


Yeah Baby!

Better yet what about syspend to disk

[Sits]

..is perfect suspend (and resume) to disk with reliable OSes. If I don't need to reboot to keep my system running why "shut it down"? Why not pause it - surely loading back a state will be far faster (if currently less reliable)?

Re:.8 sec... SO?

[marxmarv]

What I'd like to see is an EPROM to contain a kernel.

You mean like LinuxBIOS?

What is it that prevents this sort of thing?

EPROM density, tight margins, and low write speed are just three reasons. Sockets, never mind ZIF sockets, are not cheap, and don't even think about SMT sockets. Sure, you can get an 8 megabyte flash EEPROM for about $10 each by the hundred, but you're looking at a good minute or three to burn the chip full. Besides, what happens when your burn fails halfway through?

I maintain that OS'es should save as much system state across power-downs as they can, along the lines of APM sleep/wake (or better yet using an OS built out of persistent objects that can boot instantly and page in whatever is needed, on demand). Hell, with that no-POSTing BIOS and APM sleep/wake, you can already do a ten-second power-on without harming any non-volatile memory devices whatsoever. THIS IS NOT ROCKET SURGERY, FFOLKES.

-jhp

Re:Get a Mac?

[crimethink]

PowerPC systems? depends on the system. Your G3 probably uses OpenFirmware.. Macs do POST tests like any other system, and when it all checks out, you get your "Happy Mac" face, OpenFirmware jumps in, and starts your booting.

2.2 Seconds

[B.D.Mills]

Now that we've got a PIII booting in 0.8 seconds, to achieve "seven nines", we have 2.2 seconds spare. What can we do with this time? I'm sure we can do a lot of valuable system maintenanace in this time that we would not have otherwise been able to do.

We could:

• Swap out a dead power supply.
  
• Replace a faulty memory module.
  
• Swap the UPS for one that has its own in-built generator.
  
• Put fluffy dice over the console.
  
• Change the grey cables for ones in designer colours.
  
• Wave a dead chicken over the console.
  
• Upgrade the mainboard.
  
• Look busy so marketing can't have the latest item in their wish list installed.
  
• Remove the thing that marketing wanted installed because it's making the system unstable. If it's not unstable, you would not be rebooting, would you?

Of course, you might have to work fast....

Re:.8 sec... SO?

[praedor]

I have been shutting down more lately due to power costs and wanting to do my part to conserve energy. In any case, I am more interested in the laptop end of things. Virtually no one leaves a laptop up and running all the time.

It is a pain-in-the-ass chore to bootup my laptop every day. Wait and wait for the whole bootup process to finish so I can do something with it. I'd say laptops/mobiles would benefit more from an instant-on kind of setup rather than desktop systems. Also, I'd like to dedicate all the battery power to productive work on the laptop, not wasting any juice on bootup and shutdown.

Then it's not a standard C64

[Anonymous Coward]

A standard C64 takes nearly two seconds to boot up; I think it does a non-destructive memory test (SYS 64738). Unless you've got a cartridge plugged in of course.

Re:.8 sec... SO?

[mgblst]

dont work with Opera either...

are you sure your in the right place dude, doesnt work with mozilla or opera...

this is slashdot.org, not msdn.com!!

Re:Hey!

[kbeast]

get a job...

wish I had that much free time.

.kb

Re:.8 sec... SO?

[Anonymous Coward]

> One of those "Duh, why didn't we think of that before??" =)

Mind you, I thought of that before :-)

> * load simple BIOS that loads and uncompresses first xMB off the HD into ram

In my experience, reading disk with the BIOS is slower than reading with the kernel (most the goodies are not enabled when the BIOS starts).

Anyway, what I described was an implementation of your general idea. You have to take into account that the kernel would be much faster to load the disk data (do the experience with, for instance, GRUB. You'll see that it does not come anywhere near performance of UDMA). What I described was, IMHO, not difficult to implement on current hardware with the current linux boot model (looks like everything but the pre-heat stuff is already done in some patch/embedded distribution)

Note that, in your idea, the 'extracted' thing will probably have to sit in RAM forever. In mine, it'll be dealt with by the disk cache, so the RAM will be avalaible as usual for applications. It will just boot fatser, but otherwise will work as usual. Note that, even if you don't have enough ram for this root device, the system will still boot (slowly).

> Can't you postpone most of the BIOS checks to happen after the kernel has started loading?

If I am not mistaken, some BIOS test will disable interrupts, reset things, and muck with the hardware in general. Furthermore, those tests have to be done while in 16 bits mode.

Cheers,

--fred

Re:.8 sec... SO?

[Telek]

I have been shutting down more lately due to power costs and wanting to do my part to conserve energy. In any case, I am more interested in the laptop end of things. Virtually no one leaves a laptop up and running all the time.

Did you actually stop to figure out the costs of running your computer 24/7? My friend gave me the same argument, and I calculated that the computer itself, to run 24/7, amounted to about $3USD/mo. :)

As for the laptop, no, but you have 2 easy options : suspend and hibernate. Almost all laptops support both, and have for many years now, and I always just put my laptop on suspend mode, and in the unlikely event that it will be off for more than a day, or away from a power outlet for more than a day, I'll just hibernate it.

Re:.8 sec... SO?

[Telek]

last time I checked (yesterday actually) IE had a 86% market share. I'm running a Windows 2000 server. If your browsers don't implement the div tag with css style codes properly, ain't my fault =).

But in any case, I will attempt to correct the problem. Since it's just cosmetic, it's not very high on my priority list =(.

Thanks for the heads-up.

How I shortened the boot time BIG TIME.

[LF11]

I built a linux system from scratch using LFS (www.linuxfromscratch.org) a few months ago, and was quite intrigued with the boot scripts. After playing around, making custom boot scripts in perl, etc, I got a brilliant idea:

Why not run X immediately after mounting the filesystem? Thus, as soon as the kernel hands control to init, the filesystem is mounted and X is started. Then, all the daemons, network support, and big hoggy slow things start in the background (niced to 20) while I get on with my work.

Notes for those who want to do this;

Edit /etc/inittab and disable anything from using tty2. X uses it, and when getty tries to use it, getty gets the input while X keeps the display. Yuck.

Use setuidgid to start X in a user other than root.

The only real problem I've found is that I don't see errors that might occur in the boot process. (especially filesystem errors) However, once I notice something isn't right, I can Ctrl-Alt-F1 back to tty1 and see what's going on.

And, one person asked why a fast boot matters. The reason is this; most of us absolutely *hate* to wait, especially for a computer. All those little 2-, 5-, 10-second delays while the computer gets a web page, runs Netscape, whatever... they're very irritating. Like salt in a sore. Therefore, the more the waits can be reduced, the happier I am!

Yeehaw!

-lf11

Oh Great!

[Anonymous Coward]

This is a violation of my privacy!

Re:Oh Great!

[Anonymous Coward]

YA! I agree... Everyone else is complaining so why not me. Maybe I can bee cool and 3l33t like everyone else on here.

Re:99.99999%? That's great...

[CyberKnet]

yessir it did. We pulled it down, and it rebooted back up just fine. The dept wouldnt fork out cash for a new product (To track a car fleet, of all things) so they just "turned the date back ten years". Yeah. That wont hurt, right?

morons.

Re:.8 sec... SO?

[markbanang]

Why not go the whole hog and have the Kernel in microcode on the processor? Worked well enough for the Transputer. Even modern processors have a hard time competing with the context switch time of a 30MHz T800.

Take care,

Mark..........

General Software's Page for Quick Boot Soyo

[stevej_gensw]

In order to accommodate the numerous requests for more information about the General Software Quick Boot Soyo Experiment, we've set-up a web page and an email address for more details and additional direct queries. The web page contains more details, and a FAQ which the company would like to update based on inquiries to the email address. We would have liked to provide information in real time using Slashdot's software yesterday, but there were so many threads started at once, it was obvious we were going to be unable to answer every query. I would like to thank the Slashdot writers for an ongoing discussion that continues to be read by the company's employees with interest. -- Steve Jones, General Software, Inc.

Re:For those interested in this - more details

[stevej_gensw]

It took us a while to figure out how to post to Slashdot, but in the mean time set up a web page with more details. Click here to see more information about the project and what we did. Oh, we did clear up that 400Mhz thing here as well, it was indeed my fault, but what I should have said was we clocked a 600Mhz PIII at 400Mhz in safe mode. Basically, we were trying to get across a rough idea of the speed at which instructions were being retired to tie-into that 0.8 seconds. Many thanks for your post. Steve Jones, General Software, Inc.

Anyone remember FastBoot on the Amiga?

[Anonymous Coward]

Nice to hark back to the past once in a while.

FastBoot gave you a method to snapshot the entire RAM to a file then restore back from this file at boot.
(Same way that some laptops implement suspend mode I think).

From power on to full working gui around 6 seconds using an old IDE HD.
No shutdown time required as no virtual memory - just pull the plug :)

I just use 2 pcs now - one to use while the other one is rebooting.

Re:Get a Mac?

[itzdandy]

a Mac is not the answer. i agree that Macs have gone a long way in the "getting hardware reasonably up to date" department but, it seems to also take the platform longer to adopt new technologies. also, the Mac does not need a BIOS BECAUSE Macs are build identical in batches of 100's of boxes, they dont need to detect hardware in the system on each boot, its done once at the plant and burned into firmware for the next 100 boxes. you could do this on a PC but youd have to burn a new firmware for every motherboard, with everyCPU, for every type of RAM, for SCSI or IDE, etc.

Re:.8 sec... SO?

[o.sendai]

Since getting my laptop I have done less than 10 reboots, and those 10 were for installation, updates and when I forget to shut it's lid(put it to sleep) and it runs out of battery...uptimes of 40days+ so far(osx)

(Yes that is my email)

This means trouble with UID generation

[hwilker]

The Java API for java.rmi.server.UID specifies that this "UID is unique under the following conditions: a) the machine takes more than one second to reboot [...]".

Now what?