Slashdot Mirror
New Mega Alphas
GoNINzo writes: "Compaq has just announced the new Alpha servers. The have between 8 and 32 CPUs, run with a 64-bit 731 MHz Alpha chip, and current are distributed with Digital Unix or VMS. How soon before these machines are shipping with Linux preinstalled?"
More to mention is that Andrea Archangeli from SuSE was the first to get the machine booting, of course with the help of Compaq Alpha Linux experts. :-D makes thing even harder. If you buy one, ask Andrea for the patches ;-)
This machine is a new platform and you can't run alpha linux on it out of the box. The tsunami platform is the closest though. But the QBBs and the 100+ PCI busses
Care about Bogomips?
SMP: Total of 11 processors activated (14067.70 BogoMIPS).
Unfortunately with only 11 procs, but that was one of the first runs.
The natural result of that is that hacking on SMP stuff is not of top priority to the average person using Linux let alone those that are actively "kernel hacking."
That's a mouthful that effectively says "few people truly care about SMP."
Of those that do have SMP hardware, how many have more than 2 CPUs? My SMP motherboard has only slots for 2. The Slashdot What's a good motherboard for SMP Linux? discussion mostly found 2-way and 4-way SMP hardware.
Recent pricing at PriceWatch indicates that quad Xeon mobos start around $2500, and that ignores CPUs.
Certainly consistent with these being very expensive puppies that there is, resultantly, relatively little experience with.
Soup it up to 8-way SMP, and the pricing obviously heads into the stratosphere, thus further discouraging the wide deployment that allows the "open source" principle that
If it costs $20K for the motherboard, and $100K for the system, that rather diminishes the number of "eyeballs."
I think I have to agree that Linux is unlikely to be as ready to take advantage of high end SMP hardware as VMS, "whatever they want to call Ultrix today," Irix, or Solaris.
It only will get much better when there's a goodly population of kernel hackers with 16-way Alpha boxes :-). (Drool...)
Alternatively, it would be rather cool if there was some platform where we could get "massively SMP" motherboards where the CPUs didn't have to be "massively expensive." I dunno what, exactly. StrongARM would be interesting, but I gather that it is not terribly supportive of SMP. MIPS looks like the architecture for which there exist both cheap CPUs and massively parallel systems ( e.g. - the SGI/Origin "Cray" boxes).
If you're not part of the solution, you're part of the precipitate.
My zsh-powered linux seems to be more self-critical - it wants to set itself in fire, but being unable to do so, that asks for help:
zsh> So, Linux, how about that SMP support?
zsh: no matches found: support?
-- v --
SMP is working fine on sparc32 and sparc64. Haven't a clue about PPC, I think it's there but I don't recall.
The only archetecture that I know for sure doesn't SMP under linux is Mips. Mostly because Mips processors don't handle cache coherency the way other archetectures do, so a lot more motherboard logic is required to make it work. There are plenty of SMP Mips systems in existance (for instance the discontinued NEC NetPower series) but documentation for the SMP hardware implementations is scarse.
This is just like television, only you can see much further.
VMS: Rumors of my death have been greatly exaggerated.
VMS isn't dead, despite Gartner and the rest proselytizing that fate for the last 10 years. They've done the same thing to unix, until very recently.
VMS still has a large user base, but Compaq has seen fit to cut off their noses, keep prices through the stratosphere, and aim for the high end only, all the while selling off bits and pieces of technology to M$ as part of the 'Affinity' program. This was of course to woo NT development over to Alpha, but M$ as usual has found a way to make this useless to their 'partner' (delaying 64-bit NT on Alpha until Merced ships with 64-bit NT, denying any Alpha leadtime). Fortunately, Compaq has seen some of the light and pulled the plug, pulled their Digits off NT filesystem/clustering development, etc.
There are a number of organizations that will support VMS, though I bet you're more after commercial application support. There are still quite a few Apps available (even WordPerfect 7) but the market has declined into a lot of vertical applications (science, education, research, telecommunications, banking, utilities). Some of this is due to development cannibalization to port to NT, most of it is due to DEC and now Compaq treating VMS like it is a dirty word for the past 5 or 6 years.
Compaq is still a PC box pusher. They don't understand that they have an OS with incredible reliability (uptimes in the range of 13-15 years have been reported) and clustering that nothing else can touch (shared everything, over multiple transports, with automatic load balancing, cluster aliases, the ability to cluster machines 100's of miles apart, etc...)
There is still a community, despite the Q's attempts to munge DECUS into a new marketing vehicle for their desktop PC's (I get offers as a member of DECUS for steep discounts on PC's with NT installed...while what I want are steep discounts on Alpha boxes with VMS or Linux). Check out the VMSNET newsgroups or COMP.OS.VMS. Very active. DECUS has managed to get Compaq to issue a 'hobbyist' VMS license and a selection of layered products. Some commercial VMS vendors are participating to offer their products under the same license. Check out Montagar Software which distributes hobbyist licenses. You have to join DECUS (free).
The 'Open' in OpenVMS was a marketroid move when VMS fully supported POSIX (6.0?). IIRC VMS had full POSIX support before any commercial Unix did. It had nothing at all to do with the move to Alpha. It was just buzzword compliance when everything deployed had to be an 'open system'. There is no difference between VMS and OpenVMS, save for the POSIX layer. That has been removed in recent versions, as the standard VMS runtime libraries now support the POSIX API's as well.
Too bad it was before your time. When I was in college, VMS was *it*. VMS is younger than Unix, mind you, but they gave steep hardware and software discounts to colleges and phenomenal support. Now, so many useful things have been dropped from the CSLG (Campus Software License Agreement) that we plan on dropping the CSLG here next year. It is no longer worth it. Compaq's high-end blinders have lead them to sell off the layered products that made managing VMS clusters so sweet (PolyCenter Scheduler, Console Manager, Performance Analyzer, etc.) Most of these were sold to Computer Associates and now run on (and require) NT. Compaq's direction is loudly ranted about on COMP.OS.VMS.
Regardless, you could learn a few things from VMS. Linux could learn a lot, structurely. VMS was _designed_ (when Olson, the engineer, was CEO) and does so many things right... Even though Compaq has butchered the site, try the OpenVMS Website to learn more. The Documentation and FAQ links are there.
--Rubinstien
Does anyone know the current State of Alpha linux in regards to SMP. or SparcLinux or LinuxPPC for that matter. I mean I read whats in the kernel devel but I havent been able to get ahold of any real benchmarks or heard from many people running SMP systems other than on intel.
Free Unix? Free Windows. http://www.reactos.com
I saw a few of these in a Compaq SAN interoperability lab a few weeks ago. I assume they are ready now with Tru64 Unix...not sure about getting them preinstalled with linux though.
The Compaq product info page is here. The 64bit OS plus the high internal bandwith means it should be a very great clustering unit, especially as you can now fit 40 of them into a single rack.
Here are the specs:
[bash]$So, Linux, how about that SMP support?
bash: So,: command not found
What?! They still haven't intergrated the self-awareness patch into the kernel?
--GnrcMan--
Nope. Above 4 CPUs, you get diminishing returns for Linux as it stands at the moment, but it's not a hard limit (and 2.4 should do better with more CPUs, too). Take a look at http://www.dare.demon.nl/linux/sparc 64/yow.txt for an example of Linux running on a 14 CPU UltraSparc system. It's worth remembering that Linux is not just Intel. I don't think that current Intel chipsets can handle more than 8 CPUs. Machines that can take more (e.g., the Data General AV25000 can take up to 64 CPUs) tend to use multiple quad-CPU boards. These are also NUMA configurations, rather than traditional SMP.
"The invisible and the non-existent look very much alike." -- Delos B. McKown
Well the AlphaServer GS320 which is 4 clusters of 8 processors, was already shown in Geneva at the Telecom 99 show. At that time they where ironing out the multi-cluster possibilities of the system. I think this machine might have had some true potential for the Windows 2000 Datacenter, but I guess Tru64 and OpenVMS will fill that role extremely well to.
;-)
I wonder if I can cluster the 32 processor GS320 with an old 2000-300 alpha workstation ? And use a MicroVAX II as the quorum disk with VMS 5.5-H2
Uh, correct me if I am wrong, but didn't VMS die a long time ago? I remember going to it's funeral, the only thing I can conclude from this, is that Compaq == Jesus Christ if they can raise the dead like him...
Seriously, is VMS still alive and kicking? Does it have any user base, commerical support (besides compaq) or community behind it?
Differance between OpenVMS and VMS? Is OpenVMS *really* Open, or Open for marketing reasons?
Sorry, when it comes to VMS I am an idiot, it was before my time
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
75 posts and no "imagine a beowulf of these" yet ;-) Well here it is.
Spyky
Was NT supose to be Microsoft's cross-platform portable OS? I have seen stuff rolling around in old comp.sgi.* posts with people having problems getting NT on their MIPS proc... I know it isn't alpha, but the point being, wasn't one of NT goals to run on every proc know to man?
Maybe it was either a dream or an lsd trip... Mmmm roasted white rabbit never tasted so good.
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
Obviously, there aren't Alpha binaries on the W2K CD release, as MS has abandoned Alpha
Actually, you've got it backwards, Compaq abandoned MS.
--GnrcMan--
The word "open" has been used for many things. In some circles, "open systems" refers to Unix systems (as opposed to mainframe systems). These systems use relatively open standards like SCSI.
So OpenVMS runs on hardware designed for Unix systems. Old VMS ran on a Vax.
The Open Source movement is a relative newcomer to the "open" name.
So this should also explain why The Open Software Foundation, now known as The Open Group, is not an open source organization.
Analysts estimate the GS80 will cost around $100,000, the GS160 around $500,000 and the GS320 more than $1 million. The high-end servers initially will come with OpenVMS or Tru64 Unix; Linux will be available later.
:P
--- snip ---
the '80 is the 8 processor, '160 is 16, i'll leave it as an excersize for the reader to figure out the '320.
--
blue
i browse at -1 because they're funnier than you are.
Compaq says they expect $1 billion in revenue from these boxes over the next four years. They haven't named a price range, but given the market they intend to compete in, I'm guessing these are a very pricey solution for dot com web service machines... better to just get the bandwidth and a bunch of low end boxes in a cluster, I'll bet.
-- Still waiting for the Nike endorsement
The square root of everything, all news posted to slashdot is Linux VS Microsoft, please read the FAQ.
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
It was, unfortunately, a fairly typical corporate presentation, unencumbered by any real hard data.
They simulcast the CEO's announcement from NY, and we heard from another high-level manager, some dude from Oracle and a pre-recorded "howdy" from Larry Ellison. This was followed by a lunch break, some local presentations, a brief panel discussion and the requisite giveaways. Linux *did* get a mention in a couple of slides.
A great deal of time was spent talking about how great the Compaq/Oracle partnership is. I asked about Informix benchmarking, but didn't get a whole lot of response beyond "we expect the results to be good."
Whatever. There wasn't much in the way of new information. Lots of processors, Tru64 Unix scalable, manageable, really fast, etc, etc.
A brief list of buzzwords:
functionality
knowledge management
megatrend
the "edge of the web"
clickstream
"go to market"
Zero-latency-enterprise
And there you have it.
Seriously though, this is a cool system. I just wish I could afford one.
kwsNI
What the hell does Windows 98 have to do with an 8-32 processor server? Did your knee-jerk Microsoft hatred get the best of you on this one?
What the original poster was refering to is Compaq's consumer-level machines like the Presario.
These machines come with Windows98. The example was that Compaq can get Win98 for $20. If they piss Microsoft off, then maybe Compaq's cost on Win98 will rise. Even if it rose to $21, and Compaq pushed 1 million units, they would lose $1Million in profits.
Corporate interests need to be careful about every political move they make.
Having said that, Compaq already seems to be supporting Linux, so they might not be so concerned about Microsoft's agenda.
As for Beowulf, in fact the current Linux port sort of does a "beowulf-in-a-box". We support SMP up to four processors. Above that and you run multiple instances of the OS in different "partitions" of CPU-sets. Again, if a business need arose to require supporting a single instance of Linux on a 32-processor system, we could probably make that happen ("Given enough money, all problems are shallow?" 8-) )
I helped install 2 sites with GS160's in the past 6 months. There isn't a lot exciting about them. These machines are fast, because they have the new EV7 Alpha chips, but they really aren't anything groundbreaking. The same basic hardware has had 3 different names in the last 5 years TurboLaser 7000 a.k.a. Alpha 8400 a.k.a. GS160. It is a shared bus architecture that can support mulitple PCI backplanes. The thing to be on the lookout for from Compaq is called "WildFire." That will be a true distributed hardware architecture. This means Compaq will finally catch up with Sun... The operating system has a similar history (OSF1 a.k.a. Digital Unix a.k.a. Tru64 Unix.) Whenever Intel releases Merced, they have promised that the OS will support it. Keep an eye on this verison of Unix. They just added clustered filesystems, which, if it works, will make this unix ideal for distributed high-availability applications.
Call it "The operating system formerly known as the operating system formerly knows as OSF/1"
Okay, these were supposed to ship early last year, but compaq was getting lousy yields on the custom asics used in the box.
What's cool about these is they've got a crossbar switched architecture, so they scale better than a bus or omega switched network. What sucks is that the alpha architecture hasn't kept up in the clock speed race. Ghz alphas were supposed to be out by now too. Historically, the alpha has had the highest clock speed of any chip on the market. No longer. This is a bummer since they don't do as much per clock as, say, a PA-Risc system.
Still, these are sweet boxen and should deliver truly obscene tpc-c and tpc-h results. Drop a couple of storageworks coffins into the server room and all your data warehousing needs are met.
--shoeboy
(former microserf)
Right now Linux has shown only mediocre or below performance _relative_ to other, SMP optimized OS's when there are more than 4 CPU's. So you may get Linux only if you pound on the table when you ask for it, and the tech's will probably be scratching their heads in wonder that you'd spend the money for a system that Linux can't take advantage of nearly as well as DEC Unix or VMS.
Do we need to re-examine the entire approach to the kernel to really take advantage of 8+ CPU's? Why does a Linux kernel which works well with many CPU's have to be the same kernel for systems with only 1 or 2 CPU's?