Linux S5R2 App 4.21 available for Beta test

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 757052484
RAC: 1162650

RE: RE: What distro are

Message 63890 in response to message 63888

Quote:
Quote:
What distro are your boxes running?

I was just editing it...
Quote:
It can't really be the CPU type, as mine is an Athlon just like Michael's and while his does fine mine is lame without end (can't be the kernel version, either, he is running 2.6.18 and so am I). Anything else it might be... hmm... cache size? Maybe hosts with smaller case get affected worse? Dunno how much cache C2Ds have but I think it's quite a lot, and I'm quite sure Opterons have rather much... Michael's Athlon, being somewhat newer than mine, also has twice the cache. Could that be a clue?

More likely than the Distro, but I'm still not sure. I'll try to build it with the same compiler, which is the only difference I can think of that could have a large impact on the runtime. Might take a bit, though.

BM

Hi!

This sounds exactly like what we saw with app 4.17 here ==> http://einsteinathome.org/node/192684 , and also with an AMD Athlon 64. Coincidence?? Does that CPU have a auto-throttle down in case it gets too hot? Is it sensitive to the alignment of code/data for the hot-loop in physical memory? Too bad my assembler programming skills date back to the 486 :-).

[edit] BTW, the host that I'm now using for the beta app is running under Knoppix 5.0.1 (Live DVD), which is also "Debian-ish" afaik. Running fine atm, no performance probs, no SIGABRTs.

CU

BRM

Jos van Wolput
Jos van Wolput
Joined: 11 Feb 05
Posts: 47
Credit: 800840
RAC: 0

RE: Anyone else running

Message 63891 in response to message 63882

Quote:
Anyone else running Debian?


I am running Debian Sid on Intel Pentium M processor 1.70GHz (Centrino), kernel 2.6.20-1-686,
libc6 v.2.5-7.
Because half of my WU get SIGABRT with einstein version 4.18 I just installed v. 4.21

Annika
Annika
Joined: 8 Aug 06
Posts: 720
Credit: 494410
RAC: 0

Yeah, but I think we

Yeah, but I think we established it wasn't the distro. Bernd said it wasn't, his Etch box is running fine. As for thermal throttling... this CPU is liquid cooled and it never, ever got too hot, even when it was really hot outside. Yesterday we had around 14° Celsius where I live. Besides I didn't change anything except the science app and under 4.18 the box was running fine...

M. Schmitt
M. Schmitt
Joined: 27 Jun 05
Posts: 478
Credit: 15872262
RAC: 0

RE: Yeah, but I think we

Message 63893 in response to message 63892

Quote:
Yeah, but I think we established it wasn't the distro. Bernd said it wasn't, his Etch box is running fine. As for thermal throttling... this CPU is liquid cooled and it never, ever got too hot, even when it was really hot outside. Yesterday we had around 14° Celsius where I live. Besides I didn't change anything except the science app and under 4.18 the box was running fine...


Just a stupid question: Cool and Quiet ist for shure disabled?
That's the most plausible thing I can think of. I don't think distribution or cache can cause such enormous differences.
In about 5h I'll install 4.21 on my Knoppix box, as all cruching times so far look equal.

cu,
Michael

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 757052484
RAC: 1162650

RE: As for thermal

Message 63894 in response to message 63892

Quote:
As for thermal throttling... this CPU is liquid cooled and it never, ever got too hot, even when it was really hot outside.

Upps, sorry, didn't want to insult your Athlon ;-)

CU

BRM

Annika
Annika
Joined: 8 Aug 06
Posts: 720
Credit: 494410
RAC: 0

I didn't take it as an insult

I didn't take it as an insult ;-) just wanted to illustrate that scenario was highly unlikely.

Melvin Bobo Slacke
Melvin Bobo Slacke
Joined: 22 Jan 05
Posts: 32
Credit: 1808658
RAC: 74

My first ones was not 8%

My first ones was not 8% slower as indicated in the beginning but had normal runtimes.

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 757052484
RAC: 1162650

So my first beta app result

Message 63897 in response to message 63896

So my first beta app result was just validated. Runtime was not longer than for WU with same credit for 4.18.

http://einsteinathome.org/task/83943971

CU

BRM

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282700
RAC: 0

RE: So my first beta app

Message 63898 in response to message 63897

Quote:

So my first beta app result was just validated. Runtime was not longer than for WU with same credit for 4.18.

http://einsteinathome.org/task/83943971

CU

BRM

Mostly watching here, but if Annika's unit takes longer, it possibly indicates that yet again AMD systems take a hit...

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 757052484
RAC: 1162650

RE: RE: So my first beta

Message 63899 in response to message 63898

Quote:
Quote:

So my first beta app result was just validated. Runtime was not longer than for WU with same credit for 4.18.

http://einsteinathome.org/task/83943971

CU

BRM

Mostly watching here, but if Annika's unit takes longer, it possibly indicates that yet again AMD systems take a hit...

Yup, which is kind of sad. So I guess we need more beta-testers with AMD CPUs to corobborate this. Unfortunately the only AMDs I can offer are older ATHLON XPs (one Palomino & one T-bird), so far the slow down seemed to happen with the more modern Opterons and 64s, right?

CU

BRM

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.