Slashdot Mirror
2.5.65 On 32-way NUMA-Q with Preempt Enabled
_iris writes "I think the subject speaks for itself. Here is the link to the story on KernelTrap." In case you have a spare 32-processor machine munching grass in the back 40.
← Back to Stories (view on slashdot.org)
Good to see the Linux kernel making such leaps and bounds.
Keep at it guys!
I would like a ... oh wait it is :)
Rus
Cheap UK and US VPS
Nuns Under Management of Al-Qaeda
Sheesh, I'm sitting here with a 64 Way and two 32-way boxes just waiting for decent to run on them.
Does this mean that FINALLY I can shift Quake Server off the clustered S80s in the basement ?
An Eye for an Eye will make the whole world blind - Gandhi
I'd rather have a girlfriend who is also into muching carpet.
BOO! TERRO
Now this is cool. I know that SGI can scale the Altix to 64 CPU's running 2.4 with their own additions in an SSI. However not sure about. 2.5. Its nice to see it in the main kernel anyway and the only way is up
Rus
Cheap UK and US VPS
i thought i my daily 3-ways were good but it seems ive been missing out :(
Here is the complete article, the fscking lameness filter made it quite a struggle to get it posted here. Anyway:
;)
;)
;)
Zwane Mwaikambo announced today on the lkml that he's successfully boot the 2.5.65 development kernel on a 32-way NUMA-Q server with -preempt enabled. Speaking to Robert Love [interview], the kernel preemption maintainer, he began his announcement saying, "Robert, I suppose you can add another notch on your erm.. bedpost(?) and congratulations to all the kernel developers!" NUMA awareness in the scheduler was added into the development kernel in late January [story].
William Lee Irwin III [interview] explained the significance of this achievement:
"This has had a hard time historically. I'm really glad NUMA-Q's are now immune (in the sense of correctness) to this config; previously it was believed that preemption points in printk(linux_banner) would take out the machine early in boot if preemption was enabled. Congratulations rml! If you're booting without issues on these things, you are a _very_ long way toward being race-free. This is incredibly good news, both for the preemption support, and for the general stability of the i386 bootstrap."
Read on for the full thread.
From: Zwane Mwaikambo
Subject: 2.5.65-preempt booting on 32way NUMAQ
Date: Sun, 6 Apr 2003 06:48:33 -0400 (EDT)
Robert i suppose you can add another notch on your erm.. bedpost(?) and congratulations to all the kernel developers! It survived some local networking stress tests, but there is more fun stuff like tty layer to completely obliterate
(Hardware courtesy of OSDL)
Running configuration
32 Processors, PIII 500
32G RAM
Patches required:
2.5.65 (only because isp1020 decided to get huffy)
Purge assign_irq_vector panic - Zwane Mwaikambo
[boot messages]
From: Robert Love
Subject: Re: 2.5.65-preempt booting on 32way NUMAQ
Date: 06 Apr 2003 14:28:42 -0400
On Sun, 2003-04-06 at 06:48, Zwane Mwaikambo wrote:
> Robert i suppose you can add another notch on our erm.. bedpost(?)
> and congratulations to all the kernel developers! It survived some
> local networking stress tests, but there is more fun stuff like tty
> layer to completely obliterate
Excellent, Zwane.
Congratulations! Good work.
Robert Love
From: William Lee Irwin III
Subject: Re: 2.5.65-preempt booting on 32way NUMAQ
Date: Sun, 6 Apr 2003 04:23:40 -0700
On Sun, Apr 06, 2003 at 06:48:33AM -0400, Zwane Mwaikambo wrote:
> Robert i suppose you can add another notch on our erm.. bedpost(?)
> and congratulations to all the kernel developers! It survived some
> local networking stress tests, but there is more fun stuff like tty
> layer to completely obliterate
Wow!
This has had a hard time historically. I'm really glad NUMA-Q's are now immune (in the sense of correctness) to this config; previously it was
believed that preemption points in printk(linux_banner) would take out the machine early in boot if preemption was enabled.
Congratulations rml!
If you're booting without issues on these things, you are a _very_ long way toward being race-free. This is incredibly good news, both for the preemption support, and for the general stability of the i386 bootstrap.
All that's really left is driver and non-i386 arch coverage if I'm right.
-- wli
From: Zwane Mwaikambo
Subject: Re: 2.5.65-preempt booting on 32way NUMAQ
Date: Sun, 6 Apr 2003 07:25:09 -0400 (EDT)
On Sun, 6 Apr 2003, William Lee Irwin III wrote:
> This has had a hard time historically. I'm really glad NUMA-Q's are now
> immune (in the sense of correctness) to this config; previously it was
> believed that preemption points in printk(linux_banner) would take out
> the machine early in boot if preemption was enabled.
Which kernel version was that from
Microsoft just set the #2 TPC-C result in the non-clustered category using Windows Server 2003 and a 32-way Itanium 2 machine. They did this, of course, because Oracle publicly derides clustered results as not counting (and really setting up horizontally partitioned views across a huge federation of serves is not the easiest thing, and it's far from transparent for the database developer: You have to specifically design around it), so now there's a SQL Server 2000 result higher than any Oracle result.
So there you have it: A 32-way machine that's actually useful (when available on 2003-06-30).
What the hell are you guys talking about?
Yes, it is now possible to launch 32 preemptive NUMA-Q missiles strikes simultaneously using the Linux kernel. Excelent!!!
Translation: I'm going to bed, and the editors are lazy.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
I think the subject speaks for itself
If only it spoke in a language I can understand
NUMA, ok, that i understand.
(Instead of one big shared memory pool it uses processors that each have their own pool, and can access other memory with a timing penalty)
but what does "-preempt " have to do with this. what does this option do? Int unix always preemtive?
Well this is great to hear in my opinon but I'm just wondering if this will add to the Law Suit that SCO has on the plate for Linux. It's pretty sad when someone put's a lot of time and work into a project and then someone, like myself, questions if it will hurt more then help Linux. I love hearing about such innovation though and I'd like to say great job by all...
"I believe in everything in moderation. Including moderation." -Dean DeLeo, Stone Temple Pilots
IBM bought Sequent and are planning new revisions of this with a new name.
Mike Bouma
We all know that SCO invented NUMA and SMP. Jeesh.
http://saveie6.com/
Taco's Law: any story about massive scalability will be posted on a web server which craps out due to 'too many connections'.
Anyone got a mirror?
-- Ed Avis ed@membled.com
http://www.kerneltrap.org/node.php?id=628
Have fun!
Daxy's Networking Blog
Interesting that the 3 MS solutions (SQL server on Windows Server 2003) all also offer the best price/performance ratio too. Just something to think about.
Would you have all been as interested in this story if you'd known:
*sigh*
Here is a Mirror List
Mouse powered Chips, Open source Processors and Lego
Comments like this make ignorant twats like myself /. the sites, as I'm one of those who don't understand all the cryptic subjects. Please drop short descriptions, even if things seem obvious to you.
Thanks. And oh yes, it's been slashdotted already.
-a
when?
I think the subject speaks for itself.
Apparently, it doesn't.
Sometimes I worry that I'll develop Alzheimer's disease, but no one will notice.
Is that occasionally there are headlines like this that I can read, re-read, and still have no clue what the article's actually about. I don't know what ANY of that stuff means.
McDonalds makes hamburgers with the best price/performance. Just something to think about.
Would be nice if /. mirrored the stories it links to. This way only news >1 day old is accessible :(
The data /are/ safe, shithat!
Graduated out the dope game -- fatass wallets... what's that nigga's name? -- RASHEED WALLACE
IBM has been making rumblings for some time about a new 32Bit 32-Way system. The Linux kernal will be ready for it.
We're just about to mothball a 32 processor SP, maybe I'll drag it home. It could do double duty as a furnace.
Do you know what a ratio is?
Using your inexplicable analogy, that would be more like the somewhat less agreeable "McDonalds makes some of the best hamburgers money can buy, and with the best price/quality ratio". Were that true, it most certainly would be something to think about.
Or, leaving the bizarre analogy behind, if you actually look at the tables, not only does MS dominate all positions when sorted by price/performance, but in the table linked above (i.e. the best non-clustered performance) not only does MS appear three times, including the number 2 spot, showing it is extremely capable, but it also, in all thress cases, has the best price/performance of every entry.
Any more meat-patty based metaphors you'd like to use to explain that?
IBM's x440 shipping 16-way NUMA/SMP system will benefit from this significantly...this machine will also soon be going to 32-way capability...so for those of you thinking there's no hardware out there today to take advantage of this...guess again.
IBM x440
..the fact that IBM's eServer xSeries 440 is a NUMA-Q box that can scale to 16 Processors now? It is a NUMA-Q box...
As a rock-in-roll Physicist once said, No matter where you go, there you are.
It refers to preemptable work on BSD, but here is a good general description of kernel preemption.
And its a beautiful thing.
I want to delete my account but Slashdot doesn't allow it.
Its good they finally got it to boot...but still...i think there are far to many bugs in printk. I've had the 2.5's barf on me quite a bit because of this, and it only seems to get worse as it spans out over more proccessors. I think we need to proritize here. The kernel devlopers should be focusing more on stablilzing the 2.5.x kernels rather than adding loads and loads of new features. The recent benchmarks show the 2.5.x kernels are lagging way behind 2.4 and even 2.3 kernels. I think we need to stop loaded all the pretty new features for a minute and focus on getting what we have right now to work. I still have problems with ntfs writing out malformed blocks :|
:)
There is alot of cool stuff in the new 2.5.x kernels i will admit, and i look forward to using it, but as it stands i cant put a 2.5 kernel anyweres but on my home machine because once it hits a production envoirnment it craps itself. I know its just a devlopment release, but lets get it speed up a little before we start working on features for distrubted systems
I'd like to see how well a 32-way Big Mac would go down...
the kernel is getting kind of bougoise these days. With one name and all. They don't even bother to say what 2.5.65 refers too!
I guess Linux is going mainstream. Maybe I'll stop calling it RedHat or Mandrake, and just call it "2.5..."
I use Oracle all the time, and would switch to PostgreSQL before I'd go to MS SQL. But every once and a while Mr. Ellison needs a real good kick in the teeth. He's just got too much mouth for his own good.
There is nothing so silly as other peoples traditions, and nothing so sacred as our own.
...but I fail to realize the signifikance of this news item. Anyone of the in-the-know crowd: care to enlighten me? 32 bla NUMA-Q what?
-- Contradictions only exist in thought - not in reality.
2.5.65? what the hell is that?
32-way NUMA-Q? who?
Preempt Enabled? sure!
Headlines are supposed to actually make sense, fail that, the synopsis should take its place.
Ahhh, but the M$ meat patties come with gaping security holes pre-installed, whereas the others don't (or at least, probably have an order of magnitude fewer exploits running around in the wild).
No more hamburger metaphors, I just wanted to see some MS fan boy vitriol. Thanks!
Hey, maybe there is more to buying a solution than initial price/performance on a specialized benchmark test. There might be storage issues, cost of ownership, or any number of things that way well be more important in an enterprise.
Of course you can't actually buy the MS stuff until June and even then you'll be running a new hardware platform AND a new software port.
If a position in a benchmark indicates to you anything beyond doing well in one test, like being "extremely capable" I suggest that you have much lower standards for applying that phrase than most people do.
...I just wanted to see some MS fan boy vitriol....
Being an antiboy is just as pathetic than being a fanboy, and in this case he doesn't seem to be a fanboy, but you sure are an antiboy. Down with [INSERT ARCH ENEMY HERE TO BE COMICALLY AND MYOPICALLY PORTRAYED].
There might be storage issues, cost of ownership, or any number of things that way well be more important in an enterprise.
And you've added absolutely nothing relating to that, but a vague supposition that SQL Server lags. Well it doesn't anti-boy.
Of course you can't actually buy the MS stuff until June and even then you'll be running a new hardware platform AND a new software port.
SQL Server already totally dominated the clustered results, and the excellent showing in the non-clustered result is just another notch in their belt.
If a position in a benchmark indicates to you anything beyond doing well in one test, like being "extremely capable" I suggest that you have much lower standards for applying that phrase than most people do.
SQL Server is an extremely capable RDBMS. Everyone agrees with that. The only criticism that has ever held against it is that it doesn't scale as well as Oracle, and this is what is trotted out by each and every Oracle salesman and fanboy (like yourself) against SQL Server, regardless of the inapplicability to the role. Well with horizontal partitioning SQL Server now scales out pretty much infinitely, and with this result shows that it scales UP as well or more than Oracle, at a greatly reduced price. It's just one more decision point to consider.
where are all those Linux TPC results? oh, wait.. there don't seem to be any. So Linux is winning no TPC contests for either max performance or price/performance.
cpeterso
I found Unisys has some 32-way IA-32 some machines but they only support Windows (and for this kind of money the vendor is going to damn well support my OS).
Who else makes a 32-way system? Does anybody have recommendations? What do these things cost? With a good set of scripts and/or something like MPI a rack of 16 2-way servers is nearly trival to manage and utilize, so the integrated systems neeed to be around $25K to be interesting.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Storage capability seems to be a huge concern when you're talking terabytes. I honestly don't know what the picture looks like. Can I put 8 fibre channel cards in a Windows box? Maybe you can. I'm pretty interested in the Itanium2 architecture. I think it will open up the midrange. Are a lot of big sites running SQL Server? I'm sure they are. The ones at work seem to be smaller data stuff for web applications more than core large multi-terabyte stuff.
I'm sure SQL Server has a lot of features. I'm not an Oracle/DB2/whatever fanboy either; that stuff is expensive to license and difficult to administer. I know that DB2 supports a few options for stored procedures and MS does not. Oracle has very broad platform support and can be tuned extensively. I bet SQL Server is a lot easier to manage though.
I think your manner of singing MS's tune betrays your fanboy disposition. I have no loyalty to any vendor, though you're right that I don't care for Microsoft-only systems.
-Kevin
Don't click on the link in the parent post if you're at work. In fact, it's best not to click on it at all, unless you have a fetish for defecation.
There's a moderation rating for Off-Topic, but unfortunately, not for Pornography. At least the FARKers bother to label things Not Safe For Work.
Attack its weak point for massive damage!