• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Server question

D

darktiger

[H]ard|DCer of the Month - April 2007
Joined
Apr 10, 2003
Messages
11,819
Okay say I have a dual core xeon server, and I want to use more than 25% of resources for folding 1 instance, how do I do that.
 
When you start one client, it's using 100% of one core BUT widnows reports it as 25%. Reason is windows thinks 25+25+25+25=100%. The core is running at fast as it can, you just need to run more than one core.

On 8-ways, reports 13% in use... :D

Need to install more than one client. Make different folders (FAH1/FAH2 etc) then config each client with unique machine ID (1,2,3 etc).

edit: ingore HT crap. If running HT it will always report 1/2 but it's running wide open.
 
marty9876 said:
.
edit: ingore HT crap. If running HT it will always report 1/2 but it's running wide open.
Click to expand...

Ah...I don't think it works like that; pretty sure you still need to run two instances with HT. I always had to with my P4. So in the OP's case he would need 4 instances since he has 4 logical processors.
 
if you want to show 100% CPU utlization on the graph on HT based systems then yea you need to run 2x clients per physical core.

Running 2x clients in HT systems will at best yeild a 15% increase in over all points. Some WU sets you get a huge PPD hit.

Point being 2x clients on HT not needed to the one core to run ar 100% (windows reports 50% usage)
 
Roman79 said:
Ah...I don't think it works like that; pretty sure you still need to run two instances with HT. I always had to with my P4. So in the OP's case he would need 4 instances since he has 4 logical processors.
Click to expand...

No, because HT improves threading performance, but you still only have one real processor (you are still only executing one thread at a given time on an HT processor because execution bits of the CPU are not duplicated), and since it presents itself as two logical processors to your computer, Windows thinks you're only using 50% of your CPU resources when in reality you are using 100% of it. Same thing here. If you have 2 Hyperthreaded Xeons, running an instance on each should make windows report only 50% of total CPU utilization, but you're using 100%.

Correct me if I'm wrong.

I used to run 2 instances on my P4C, but it slowed everything waaaaay down.
 
marty9876 said:
if you want to show 100% CPU utlization on the graph on HT based systems then yea you need to run 2x clients per physical core.

Running 2x clients in HT systems will at best yeild a 15% increase in over all points. Some WU sets you get a huge PPD hit.

Point being 2x clients on HT not needed to the one core to run ar 100% (windows reports 50% usage)
Click to expand...

I think your right for xeons, but these dual cores do way better than 15%...thats why on a good day you'll see 2000 plus ppd's
 
Differance is HT is not a core.

Dual core= 2x clients per physcial CPU.

HT= 1x client and windows reports 50% usage

Dual core HT Xeon chips= 2x clients

Ahh I give up. This coversation has been had so many dimes. :)
 
marty9876 said:
Differance is HT is not a core.

Dual core= 2x clients per physcial CPU.

HT= 1x client and windows reports 50% usage

Dual core HT Xeon chips= 2x clients

Ahh I give up. This coversation has been had so many dimes. :)
Click to expand...

You're still wasting parts of the pipeline if you don't run 4 clients on a dual HT xeon box. I run 4 on my dual prestonia, ever bit helps.
 
ND40oz said:
You're still wasting parts of the pipeline if you don't run 4 clients on a dual HT xeon box. I run 4 on my dual prestonia, ever bit helps.
Click to expand...

Not really.

HT boxes are sharing parts of the CPU of which the CPU's have only one of. If your running 2x (4x) of the same exact WU's which dig into the SSE instruction sets (GROMACS) you not really gaining anything.

Now if you are running 1x GROMACS and 1x Amber, then you can have a nice increase. It really all depends on the types of cores running.

OC-AMD per Larry said:
Folks, for many the HT (HyperThreading) issue is one that they just don't understand. While Stanford allows you to run more than one instance of the Folding@Home client on your computer, you are not really helping them if the number of clients you are running is not equal to the number of REAL CPU's you actually have. While running two instances of the F@H client will return two work units more quickly than doing one at a time, that is not what is important. Returning one work unit 70% faster and starting the next generation of tests on the work unit, is what Stanford wants. Speed in working through each protein generation is very, very important. Sometimes quality is better than quantity. Quality in this case is work being returned more quickly. While it is well known that an HT CPU can run 2 work units because the OS treats it like two CPU's, and you can gain a possible 15-30 percent increase in points for that computer, it also means that each work unit is returned more slowly. In simple terms, if the project has 300 generations needed to test a theory on a protein model, and running two instances at once delays the return of work for 1 day each time, you end up with a 300 day delay. That translates to about a 10 month delay in examining your data for the final result. The bottom line is simply this: Run 1 instance for each CPU you have. An HT CPU is not two CPU's, it is one. Let's work to advance the science and spend a little less time worrying about the number of points you get.

This is a direct quote from Dr. Vijay Pande:
1) If you care primarily about points, running 2 procs on HT is still the best bet. We are grateful for all contributions and if people choose to run 2 procs on HT, our approach is that all contributions are welcome.

2) If you care about the science foremost and are interested in our recommendations, then do not run 2 procs on HT, but please just run one process. That won't be best for points, but is best for the science.

3) If your machine cannot make the deadlines, then one should run the timeless WUs.

I hope that clears these issues up and thanks to all for their contributions.

Thanks for listening and please do the right thing.
Larry
Click to expand...
 
From my experience Marty and ND40oz are both right. With a dual core xeon I'm assuming you have two dual core chips, so 4 cores total. If you run 1 instance it will be reported as 25%. The only way to get higher is to install more clients and set their cpu priority in the advanced setting of the install.

If you have an old style xeon setup with 2 HT chips then you can just intstall 2 clients and windows will report 50% when it is really using 100%, Now you can squeeze a little bit extra out by installing an additional 2 clients(total 4), but you risk super saturating your memory bandwidth ect. and overly slowing things down.

 
You must log in or register to reply here.
Back
Top