• 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.

For the [H]orde

P

phorensic

n00b
Joined
Jun 18, 2003
Messages
61
Added some new clients this week.

fahclients-1.jpg

fahclients-2.jpg


I think I have about 77 clients now. FahMon says 20Kppd now, but we will see. We recently moved to Citrix. Had to image all our desktops. Gave me the opportunity to install FAH on a large chunk of machines that are now just sitting there idling away running nothing but the low demand Citrix client.

Isn't it crazy how many standard CPU clients it takes me to hit 20Kppd? Wish I had some nVidia GPU's above 8000 series around here.
 
phorensic said:
Isn't it crazy how many standard CPU clients it takes me to hit 20Kppd? Wish I had some nVidia GPU's above 8000 series around here.
Click to expand...

I think that sentiment is shared among a lot of folders. As long as we keep using the uniprocessor it will never change though. Maybe it will never change no matter what. I really would like to know honestly what PG thinks of the uniprocessor and the work it puts out. If you judge based on points it would tell you they think very little of it.

Still, I commend you for borging those computers anyway. I hope you have permission to do so. I would hate to see a fellow [H]orde member getting in trouble about such things again.

Keep it up!



 
Nice setup. same sentiment on the permission question.

 
Wheresatom said:
I think that sentiment is shared among a lot of folders. As long as we keep using the uniprocessor it will never change though. Maybe it will never change no matter what. I really would like to know honestly what PG thinks of the uniprocessor and the work it puts out. If you judge based on points it would tell you they think very little of it.

Still, I commend you for borging those computers anyway. I hope you have permission to do so. I would hate to see a fellow [H]orde member getting in trouble about such things again.

Keep it up!



Click to expand...

Yeah, 1 GPU at home on that list and a few SMP clients almost completely match the rest of the single core clients. Crazy.

I have permission from the IT guy above me, but corporate probably wouldn't be too happy about it. It does not affect the end user's experience whatsoever, as it is an idle whoring app as we know, and these machines have plenty of memory to let it crunch away unnoticed. Also, the upload/download of units is capped at 280Kbps or our 10Mbps WAN connection, so it can hardly affect our WAN. HOWEVER, the electricity drain is probably astronomical spread across all those machines. It's a fun "stealth" project. I have learned a metric shit-ton about running and managing programs remotely on machines doing this.
 
phorensic said:
I have permission from the IT guy above me, but corporate probably wouldn't be too happy about it.

<snip>

It's a fun "stealth" project
Click to expand...

Two things, and I'm not trying to be an asshole here but.....

The IT guy doesn't own the machines, corporate does.

Corporate probably doesn't really support "stealth" projects, and as said above, I too would hate to see a fellow folder in trouble....

Regardless, I am one of those still running a borg that consists of nothing but the standard client, and the way I look at is this.... If the work didn't need to be done, Stanford wouldn't be sending the wu's out... right? They might not be "worth" much point wise, but the science needs to be done, and I will continue to keep my borg alive until they no longer get work...

And on top of that, you can't beat the reliability of that client, I check in on my borg maybe once a month at best, and as long as no hardware has died, the damn things are still running 100% reliable..

Beers and tamales, and Fold On!!!

 
+1 for digital here. Even though the PG doesn't want to value the standard client with more points (hopefully they will at some point), I still think there is considerable value in the work they produce. As for reliability when I got home last night all 5 of my SMP clients were stopped at 100% not able to shut down and send in results. That kind of crap just doesn't happen with the standard clients. Any way you look at it 77 clients is a lot of work. Nice job. :cool:

Folding is never worth your job so make sure you are covered.

 
To elaborate on what digital and highyield have already said, check out the two links in the [H]orde Central thread about woods from OCF and the Horde's own Party2Go. They thought they had the right permission. We just don't want to see anyone taking even the slightest risk.
 
Nice borg.

It's nice that you can get all of them on FAHMon too.

 
I understand your concerns and still it's a risk I'm willing to take. I'm 25, I make shit for pay in this position, I don't have a wife, kids, or a house to lose, so if they want to fire me for trying to help humanity with this research project for Stanford University, they can. Especially since I have learned so much about remote process/program management, which relates HEAVILY to our company and the position I am in. As for points zeroing..geez...if someone wants to call me out for helping humanity (again), go ahead. Funny stigmas surrounding this project. I have permission from the guy above me, and while he may not be the legal owner of the machine's, he purchased them directly, he manages me, and that is the way things work in our region. I don't approve every little thing through corporate, nor should I...Hell, half the machines are out of warranty and useless to corporate anyway (which was a recent announcement here). Interesting, though, how people deal with this. Also, my boss is curious as to how the whole folding project works...and people walk up to my cube and see my naked Athlon 3200 machine folding away, get curious, I explain the folding project, and they seem genuinely interested and thank me for contributing to the cause.

@Kendrak -> FahMon constantly screws up it's database of clients. It's a pain to keep them all active in FahMon without the program crashing or changing the file paths. However, out of all the monitoring programs, I have had the best luck with it, so I continue to endure the pain of using it.


 
phorensic said:
I understand your concerns and still it's a risk I'm willing to take. I'm 25, I make shit for pay in this position, I don't have a wife, kids, or a house to lose, so if they want to fire me for trying to help humanity with this research project for Stanford University, they can. Especially since I have learned so much about remote process/program management, which relates HEAVILY to our company and the position I am in. As for points zeroing..geez...if someone wants to call me out for helping humanity (again), go ahead. Funny stigmas surrounding this project. I have permission from the guy above me, and while he may not be the legal owner of the machine's, he purchased them directly, he manages me, and that is the way things work in our region. I don't approve every little thing through corporate, nor should I...Hell, half the machines are out of warranty and useless to corporate anyway (which was a recent announcement here). Interesting, though, how people deal with this. Also, my boss is curious as to how the whole folding project works...and people walk up to my cube and see my naked Athlon 3200 machine folding away, get curious, I explain the folding project, and they seem genuinely interested and thank me for contributing to the cause.

Click to expand...

In that case, no problem :cool: Some people isn't aware of them and bitch when shit happen to them so we wanted to warn you about that but since you seems to be fine with the consequences, it's fine for us as well.

Fold on !!

 
Local charities solicit all the time. Companies help for both tax reasons and PR.
Might be worth making up some print outs with F@H information for those stopping by to see your "naked Athlon 3200 machine folding away".
Your boss would be able to better explain to his how your dept/company is helping find a cure pre-armed with a print out detailing Folding @ Home.
Might be worth stopping by HR as they are those approached when asking for donations.

Hate to see you in trouble for helping. :(
 
I asked IT at my company about folding on my own machine. It wasn't a very enjoyable experience. I am not IT so I only use one comp, a Dell GX620. They had no idea what it was and they talked about it in meetings and then decided they weren't smart enough to decide if it was safe or not. I gave them copies of the logs so they could see what it does when it accesses the servers at Stanford.

They all said they understood that it was for a good cause but our "Computing resources" were not to be used for non-business purposes. Then finally a month later my boss asked me if I had used company time relating to the project.

All this for one Pentium D that never does anything more intense than printing PDFs to an outdated HP network printer. The moral of this story is that any decent sized company is liable to have some paranoid people that would find some way to condemn your efforts. So be careful with those borgs.

 
phorensic said:
Isn't it crazy how many standard CPU clients it takes me to hit 20Kppd? Wish I had some nVidia GPU's above 8000 series around here.
Click to expand...
Indeed, that's the chief reason why some huge producers like FLECOM largely abandoned F@H and maintain only a nominal number of clients at present. For those with access to a large farm, it's becoming impractical for many reasons to run the standard client alone. Besides the dearth in PPD compared to the high performance clients, there's also the question of power consumption vs science accomplished. I have only one uniprocessor box left after converting another one to a pure GPU machine. By the end of the year, the only standard clients I'll have left are the ones I have running solely to utilize spare cycles my virtual machines leave unexploited.

BTW, are you running the uniprocessor client with standard WUs or bonus WUs?

 
Most are 6.22 run in uniprocessor mode with [clienttype] type=3, which is:
S: type=3
O: Request work units without deadlines (no/yes) [no]?
O: Set -advmethods flag always, requesting new advanced
scientific cores and/or work units if available (no/yes) [yes]

This nets me max PPD per physical CPU. I have a mix of P4's without HT, P4's with HT, Core2Duo's, and one nVidia 9600GT. I learned recently that you should run only one client on a P4 with HT, as the 50% usage in task manager is incorrect due to the way HT works. I was previously running two clients per P4 w/HT. I am running 2 clients per Core2Duo, because it is a major major major pain to try and manage an SMP client remotely on a domain machine (especially when the end user logs off). My SMP machines are off the domain or at home. My GPU client is on my home machine.
 
Right now I don't think that WU size matters, its -advmethods (aka type=3) that gets you the good stuff.
 
You must log in or register to reply here.
Back
Top