|
[Sponsors] |
July 14, 2006, 11:55 |
Re: Woodcrest benchmark data
|
#21 |
Guest
Posts: n/a
|
Just looked around. It seems woodcrest is just 5100 series Core 2 Duo Xeon.
I found DELL Precision 490 or 690, HP xw8400 have this kind of CPUs. Are they "woodcrest" workstations? Did they eliminate the memory bottle neck problem for regular dual intel P4, Xeon CPUs? If the answer is yes and yes, I would think to buy them to set up my cluster. |
|
July 16, 2006, 17:41 |
Re: Woodcrest benchmark data
|
#22 |
Guest
Posts: n/a
|
Hi,
To my knowledge you are correct, the Intel 5100 series processors are the Woodcrest ones. I have checked with Dell and they have not put Woodcrest CPUs on the web site yet (at least here in Australia but we are always behind the rest of the world) but they are available and will go up shortly. In the Dell server range ask about the Dell 2950 series servers. Also have a look at their benchmark results on www.spec.org - quite impressive - pretty much as good as the top itanium CPUs per processor and a fraction of the price. Also keep in mind that it appears the 5100 series CPUs are interchangable with the 5000 series CPUs. The 5000 series are not Woodcrests and are considerably slower. Make sure you specify 5100 series CPUs. Regards, Glenn |
|
July 16, 2006, 17:45 |
Re: Woodcrest benchmark data
|
#23 |
Guest
Posts: n/a
|
Hi,
I think there is a lot more to it than just the cache size. These processors are based on a entirely different architecture than the previous generations of workstation/server CPUs. For starters the fact that Stumpy reports dual core operation to be just about twice as fast as single core operation indicates Intel have improved the CPU to memory bottleneck which every previous Intel 32 bit CPU had previously. Then you factor in support for FB-DIMM memory, 1333MHz FSB etc etc and there are plenty of differences. Have a look at the Intel web site if you want the full schpeil. Glenn Horrocks |
|
July 16, 2006, 20:02 |
Re: Woodcrest benchmark data
|
#24 |
Guest
Posts: n/a
|
Hi, Glenn,
Thanks for your info. Now I know DELL also sells computers in Australia. If you can check U.S. DELL small bussiness section, the 5100 series CPU is available on Precision 490 and 690 with much high price than 5000 series. I have configued a DELL 690 with two 5160 (2.66 GHz), 16 GB 677 MHz RAM. The price ($6200) is a little over than four DELL 380 (3.4GHz P4 CPU, 4 GB RAM for each). I am waiting for the money to buy the DELL 380. If the woodcrest 2.66 GHz CPU can beat P4 3.4 GHz CPU, I will go for woodcrest. Someone reported 3.0 GHz woodcrest took half of the time on the benchmark problem. If this is true, woodcrest CPU would be the one. |
|
July 17, 2006, 04:10 |
Re: Woodcrest benchmark data
|
#25 |
Guest
Posts: n/a
|
Hi,
For sure there are more facts affecting these results Patrick |
|
July 17, 2006, 17:28 |
Re: Woodcrest benchmark data
|
#26 |
Guest
Posts: n/a
|
Hi,
Based on the results presented here it looks like the Woodcrest option you mention would be twice as fast as the P4 option. If the price is similar the Woodcrest sounds like better value for money. Does Dell supply the FB-DIMM memory? The quote I got from Dell used DDR2 RAM. Any idea whether going to FB-DIMM memory would be worth it? You might also consider going for the faster 3.0GHz processor as it is cheaper to buy extra speed as a faster CPU rather than extra parallel licenses. We can always find wonderful ways of making the purchase more expensive. Glenn Horrocks |
|
July 18, 2006, 09:44 |
Re: Woodcrest benchmark data
|
#27 |
Guest
Posts: n/a
|
There should be more feedbacks from users regarding CFX speed on Woodcrest CPU.
DELL does supply FB-Dimm. Precision 490 and 690 use DDR2 SDRAM FBD. Bian |
|
July 18, 2006, 22:56 |
Re: Woodcrest benchmark data
|
#28 |
Guest
Posts: n/a
|
Hi,
Intel has been also making noise about the new Conroe CPUs, which are meant to be the desktop/workstation version of the Woodcrest (Woodcrest was aimed at servers). Does Conroe provide similar performance? I think they are still a few weeks off being available. I would assume it is potentially a little better still. Regards, Glenn |
|
July 19, 2006, 05:02 |
Re: Woodcrest benchmark data
|
#29 |
Guest
Posts: n/a
|
Woodcrest and Conroe have a common core hence performance at identical clockrates will be virtually identical.
Of course the slightly higher bus speed and use of a quad channel memory configuration gives Woodcrest a slight edge. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Problem in running ICEM grid in Openfoam | Tarak | OpenFOAM | 6 | September 9, 2011 17:51 |
Benchmark data for temperature field in 2D cavity flow | Vladislav | Main CFD Forum | 0 | June 18, 2010 12:33 |
benchmark data | Zarchary | Main CFD Forum | 0 | March 23, 2007 17:30 |
Benchmark flows? DNS data ? | karuna | Main CFD Forum | 0 | June 3, 2006 18:05 |
How to update polyPatchbs localPoints | liu | OpenFOAM Running, Solving & CFD | 6 | December 30, 2005 17:27 |