[BOINC] [生命科学类] PS3GRID & GPUGRID

Youth

正如项目名称，这个BOINC项目是设计运行在PS3上面的。

http://www.ps3grid.net/

PS3GRID is a volunteer computing project based on BOINC for full-atom molecular dynamics simulations and other scientific applications specially optimized for the Cell processor and the PlayStation3.

如果你有PS3，可以发邮件至ps3grid@gmail.com来申请帐号。

tyrocn

呵呵，那就只能纯支持了

zglloo

除了支持没有什么好说的

MythCreator

为什么没有For X360的项目呢

cnchina

有FOR CELLPHONE項目么？

Youth

呵呵，移动设置还是不要算了吧，不然电池更不够用了。。。：）

universebreaker

原帖由 MythCreator 于 2007-6-5 11:46 发表
为什么没有For X360的项目呢

臭bill會給我們用x360算包的一天
就是另一個新時代的開始
也是全球歡騰的一天
因為那意味著臭bill的獨裁皇朝結束了

Youth

Finally GPU computing on BOINC! After delivering BOINC on Playstation last year, we are happy to announce that from today we are are able to distribute workunits on Nvidia GPUs. PS3grid.net and the new website GPUGRID.net go along to create a large infrastructure for biomolecular simulations. To run GPU workunits simply follow the instructions on the website. Apart from us, this was made possible by a strong collaboration with BOINC developers and in particular David Anderson at Berkeley and Stefan and Keith from the BOINC community, already moderators at ps3grid. We also want to thank Nvidia for their important support.

开始支持GPU计算！

应该只是在原项目中新增一种计算平台，不清楚为啥要弄两个域名，可能是怕ps3grid的名称有误导作用吧。。。

目前只支持64位linux平台，关于windows平台，可能要一个月后了：

[quote]What about Windows support?
I am not a linux person and don't even know how to setup boinc in linux

We will support Windows but I don't know when. It could be in month or later.

gdf
[/quote]

Youth

But I received two different work unit, the second was longer than the first one with same credit. (16 vs 18 hours)

So 112 - 125 points per hour with 9600GT @ stock // E4300 @ 3ghz // Ubuntu64 hardy // forceware 173.14

ZeuZ @ L'Alliance Francophone

9600GT，第小时得分率112-125。

///

I got successful runs since a few weeks on the hidden test project... ;)

http://www.ps3grid.net/show_host_detail.php?hostid=4674 has a Q6600 @ 2.63 Ghz, 4GB DDR2 800, a EVGA 9800 GTX SC and is running Fedora 9 64bit with the 173.14.09 nvidia drivers .
Runtimes are about 35.000 seconds per WU...

http://www.ps3grid.net/show_host_detail.php?hostid=4675 has a Q6600 @ 2.7 Ghz, 2GB DDR2 800, a Gainward 8800 GT and is running Ubuntu 7.10 64bit with 169.12 nvidia drivers.
Runtimes are about 45.000 seconds per WU...

And I'm thinking about to buy parts for a new computer with Q9300 and a new GTX 260 craphics card which should be a pretty nice GPU for crunching... Costs for all parts would be about €700.-

9800 GTX上大概35000秒一个任务包，得分1900+，同样的任务包在8800GT上大概45000秒。

///

GPU飙分的效果不错啊~~不知道GPU计算时对CPU的使用情况怎么样。

Youth

[quote]Thanks guys!
What I would like to know, is how CPU intensive is this? I am imagining the GPU is doing most the work. A core is tied up with it but not running 100% like we do with "traditional" projects.

Actually the application uses also one core of the CPU to 100% plus the GPU as coprocessor...
I really can't say how important the CPU speed is for the GPU application, but earlier test have shown that if the app uses only 50% of a CPU core, the WUs were a good bit slower.

Would be interesting to see some comparisons with the same graphics cards but other CPUs in the thread you started...[/quote]

The CPU is not important at all. It appears to be using 100% of resources just because it is polling waiting for the accelerated kernel to finish.
So, any CPU should do.

We have built a machine with 6 GPU cores by putting together 3 Geforce 9800 GX2, a power supply of 1200W, an Nvidia 780i motherboard with 3 PCI-E 16x slots and a quad core CPU.

We are looking into building another one with Geforce 280.

说是CPU完全不重要，看来可以CPU+GPU全速飙了~~

// 也可能是说总归要占用一个CPU，而不管那个CPU是好是差。。。

Youth

After few days of testing, we are pretty happy with the results. Only minor issues.

Please help us to tell the BOINC community via your teams, news sites and son on.

GPU computing on BOINC is step change for computational science.

GDF

经过几天的测试，我们对结果很满意，只有些小的问题。

请通过你们的团队、新闻站点等将我们的情况告诉整个BOINC社区，BOINC平台上的GPU计算是计算科学的一大进步！

//

只有差显卡的俺郁闷地飘过。。。

Youth

24 Jul 2008 9:54:04 UTC

Hi,

Several of you have asked exactly which graphics cards are supported by GPUGRID. All Geforce 8, 9 and 200 series GPUs are capable of running CUDA code. In practice, however, only those with 96 or more stream processors are fast enough to participate usefully. In practice, this means a 8800, 9800 or GTX260/280 card.

既然大家在问GPUGRID支持哪些显卡，理论上，所有支持CUDA的8、9、200系显卡都是可以的，但实际上，显卡至少是有96个或更多的流处理器，计算起来才比较有效率，这意味着得是8800、9800或GTX260/280这些卡才行。

You may look up the configuration of a particular card on the following wikipedia pages:

• Geforce 8 Series Cards
• Geforce 9 Series Cards
• Geforce 200 Series Cards

各系列显卡的详细规则可以查看wikipedia上的介绍。

Currently we use 8800GT and 9800GX2 cards for our development.

目前，我们用来开发的是8800GT和9800GX2。

Note that because the GPUGRID code works the graphics card very hard, we do not recommend using an overclocked card.

请注意，因为GPUGRID对显卡的使用强度非常大，建议大家不要超频使用。

MJH

DrLingCN

啊～～～～～～～～～～～～
我的8600M GT不行？！效率低？！

Youth

嗯，能算，但应该效率比较低，我刚查了一下，这个卡是32个流处理器吧，至少比我的8400M GS好多了：）

DrLingCN

最多只有8800的1/4到1/3的速度……