Message boards :
Number crunching :
Show and tell your machine. Here's mine.
Message board moderation
Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · 9 · 10 . . . 20 · Next
Author | Message |
---|---|
ivan Send message Joined: 5 Mar 01 Posts: 783 Credit: 348,560,338 RAC: 223 |
Similar spec to my new server -- I got one half of it running today. Here's a view from the back showing the two redundant power supplies in the middle and one of the server slides. Hasn't got any GPUs (yet... :-) but there's 128 GB of RAM in each server. Actually, I am running a 12-core machine without Hyperthreading (it's the server for my Xeon Phi, which is unfortunately unsuited to s@h). Its RAC is about 30% less than yours, which is probably partly explained by the 17% lower clock but also by Linux/Windows differences. With two identical machines, down to the smallest detail, running one with and one without HT perhaps I can convince myself which way is best. At the moment, the second machine is one day younger and a bit behind -- but it managed to grab a large bunch of AP jobs and at one point was crunching 18 of them at once; at 15 hours or so each and 14 still pending, that left a bit of a hole in its earnings so far. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Similar spec to my new server -- I got one half of it running today. Here's a view from the back showing the two redundant power supplies in the middle and one of the server slides. Hasn't got any GPUs (yet... :-) but there's 128 GB of RAM in each server. I have 5 identical i7-860's and sometimes I see as much as 15% difference between them. Depending on the work they are processing. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
ML1 Send message Joined: 25 Nov 01 Posts: 21129 Credit: 7,508,002 RAC: 20 |
... I turned hyperthreading off to see what the difference might amount to. Let us know for what you find for the 'hyperthreading'. For my own applications, I've been very underwhelmed with the throughput and I'm left wondering if the whole idea is a x2 Marketing ploy for a meagre x1.1 performance... Happy fast crunchin', Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
James Sotherden Send message Joined: 16 May 99 Posts: 10436 Credit: 110,373,059 RAC: 54 |
... I turned hyperthreading off to see what the difference might amount to. On my I7 920 running stock It was a close call. I didnt see a whole lot of differance. With running lunatics I dont know. After almost 6 months of running stock doing all sorts of tests Im not going there:) But leaving a free core to feed a GPU could be true. Just marginally on this compter. But for any others YMMV. [/quote] Old James |
betreger Send message Joined: 29 Jun 99 Posts: 11414 Credit: 29,581,041 RAC: 66 |
On my I7 920 running stock It was a close call. I didnt see a whole lot of differance. With running lunatics I dont know. After almost 6 months of running stock doing all sorts of tests Im not going there:) But leaving a free core to feed a GPU could be true. Just marginally on this compter. But for any others YMMV. Good responce, makes sense to me. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
... I turned hyperthreading off to see what the difference might amount to. When Intel first released HT I think their statement was something along the lines of 15-20% gains could be expected from the use of HT. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
... I turned hyperthreading off to see what the difference might amount to. In the early days of HT, I did some extensive testing and the results were very underwhelming. I finally found that running HT off was more stable when OCing and ran cooler. I have not used HT since. "Time is simply the mechanism that keeps everything from happening all at once." |
ivan Send message Joined: 5 Mar 01 Posts: 783 Credit: 348,560,338 RAC: 223 |
... I turned hyperthreading off to see what the difference might amount to. My leaning is not to run HT as these are research machines and we are more interested in single-thread efficiency than overall throughput; but, I'll run like this for some time if I can. It's not often I get to play with two identical machines simultaneously. |
ML1 Send message Joined: 25 Nov 01 Posts: 21129 Credit: 7,508,002 RAC: 20 |
I guess this would be the place to post the obituary. Dare I share this link to tickle your whiskers?... AMD claims record with latest overclock-happy FX Series chips, again How's 8.7GHz at 125 watts sound? Plus, eight cores at 95 watts... Note: Even at such a phenomenal overclock, you should still get better by overclocking a GPGPU... Happy super cool super fast crunchin'! Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
ML1 Send message Joined: 25 Nov 01 Posts: 21129 Credit: 7,508,002 RAC: 20 |
... My leaning is not to run HT as these are research machines and we are more interested in single-thread efficiency than overall throughput; but, I'll run like this for some time if I can. It's not often I get to play with two identical machines simultaneously. I've been trialling a HT enabled CPU for some time now and I've found that the HT is of too little gain for my application to be worth the Intel expense. You also get a repeated surprise when you are under the impression that you have for example 4 CPU cores to use to then find they in effect run at half speed when you get HT tasks sharing the same physical core. Completely fouls up job execution time estimates! Hence, I'm now minded that unless you really do need CPU-FPU number crunching, then AMD seem to have the better mix of two cores sharing a "HT"-style FPU. Also, that is all at much lower hardware cost. So... To get back to s@h, go AMD + multiple-GPUs for a compute monster?! ;-) Happy fast crunchin', Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
James Sotherden Send message Joined: 16 May 99 Posts: 10436 Credit: 110,373,059 RAC: 54 |
As for running Intel chips and Nvidia cards. I believe its what you have installed, That could make a differance. On my I7 920 host with a GTX 250 I did not see much differance running HT on or off. That was with running stock MB only though. Im now running the new lunatics app on that and my older I7 3770 with a GTS 550 Ti. I think when the new AP V7 comes out, I might try a doing HT On and OFF both of those to see what happens. I realize that the GTX 250 is way old and the GTS 550 Ti is now long in the tooth. But A lot of folks still run them. [/quote] Old James |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
As for running Intel chips and Nvidia cards. I believe its what you have installed, That could make a differance. On my I7 920 host with a GTX 250 I did not see much differance running HT on or off. That was with running stock MB only though. The advantages/disadvantages of HT are probably down to the generation of hardware. I think newer hardware probably show more usefulness than previous generations. Such as my i7-860 producing 28% gain in output with 11% more power consumption while doing MB work. Where your i7-920 only showed 10% gain in output with HT on for MB work. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
As for running Intel chips and Nvidia cards. I believe its what you have installed, That could make a differance. On my I7 920 host with a GTX 250 I did not see much differance running HT on or off. That was with running stock MB only though. The i7-860 and i7-920 are the same generation CPU; they are both first gen Core i products. Second generation prefixed a 2 in front of the model, such as Core i7-2600. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
As for running Intel chips and Nvidia cards. I believe its what you have installed, That could make a differance. On my I7 920 host with a GTX 250 I did not see much differance running HT on or off. That was with running stock MB only though. They are both first gen but a bit different from one another. Bloomfield vs Lynnfield. To me it would be a bit like comparing an i7-4820K with an i7-4770K because they are both 4th generation i7's. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
As for running Intel chips and Nvidia cards. I believe its what you have installed, That could make a differance. On my I7 920 host with a GTX 250 I did not see much differance running HT on or off. That was with running stock MB only though. Ah, I see what you're saying now. The core architecture is indeed different. Calling them a different "generation" though threw me off a bit. In Intel parlance they are both Nehalem micro-architecture CPUs, but with different core revisions. |
Michel Makhlouta Send message Joined: 21 Dec 03 Posts: 169 Credit: 41,799,743 RAC: 0 |
Here's my main cruncher, 7256386, running mostly 24/7 except while I am gaming. It has a 4770K with an MSI 780 (will add another whenever I can find one where I live), same cruncher related to the 4770K overheating thread: I've also got a couple of laptops running seti as well. Dell XPS 1645, 7046957, with a spare fan to keep the noise and temps down: Dell Inspiron 9300, 7172439, this one is ancient that I've revived a year ago |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
My latest PC, a HP Pavilion 500-152ea is simply a black box, so there is no need to show pictures. But I think its 2 TB disk, placed vertically and bolted to a chassis, is a Seagate hybrid disk, with a SSD cache of 8 GB. The HP datasheet of a similar box, the Pavilion 500-240el, says it has a 16 GB SSD cache, but the datasheet does snot specify the make. Anyway my new PC runs well with its AMD A10-6700 CPU at 3.7 GHz and produces less heat than my SUN WS of 2008 vintage and my HP 635 of 2012 with its AMD E-450 CPU at 1.6 GHz. The SUN has a very reliable Opteron 1210 at 1.8 GHz. 8 GB RAM on all three, DDR2 on the SUN and DDR3 on the HPs. Tullio |
Fred J. Verster Send message Joined: 21 Apr 04 Posts: 3252 Credit: 31,903,643 RAC: 0 |
I use one self build computer. In an COOLER MASTER Case [Older type] An INTEL dp 67 BG Mobo, 1 I7-2600 CPU, 2 x 8GByte DDR III 1600MHz. DRAM. Two AMD/ATI HD5870 GPU 's and a 1000Watt P.S.U. {Using ~455Watt/h crunching (HT=ON --> 1,4 more comp.power).The DDR3 DRAM is faster and has lower latency. I use a stock - cooler which can't handle the heat from the CPU, so it's on 85% , CPU just sits below 100C . . .GPU 's are 20 % used (EINSTEIN, 2 WU's per GPU). Pretty 'Basic Set Up' in such case. But works as it should. |
arkayn Send message Joined: 14 May 99 Posts: 4438 Credit: 55,006,323 RAC: 0 |
I use one self build computer. In an COOLER MASTER Case [Older type] A quick after shot. |
Wiggo Send message Joined: 24 Jan 00 Posts: 36608 Credit: 261,360,520 RAC: 489 |
OK I've been asked to do this enough times now so after removing the winter dust bunnies I broke the camera out before I sealed them up again. The 3570K with dual GTX 660's. And the 2500K with dual GTX 550Ti's and GTX 560. Cheers. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.