• 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.

ML350 G5 bigadv capable

nwrtarget

Gawd
Joined
Aug 10, 2010
Messages
903
Hey there wise ones.
I have an ML350 G5 sitting here with dual quad core 2.0 GHZ CPU's. It has 2 x 512 and 2 x 2 gig sticks in it and is running CentOS 5.

In your opinion will it make the quick return bonus on Bigadv?
 
I don't remember exactly which Quads HP used in that but I don't believe they have HT so you will only have 8 threads at 2.0 GHz. I don't believe it will be fast enough mate.
 
Uncle Tobit is right no HT on this server.

So he thinks it isn't fast enough. Any other opinion providers?
 
I'll bet it will be close. Post some smp times from it. That should give is a very good idea. Veeb0rg has an 2 x 4 dually with 8 cores at a slower speed that I never would have though would make the deadline, but does easily. Native Linux w/ BFS will hal a lot.
 
It is running CentOS 5.6 and needs to stay that way. I won't have it very long so I was just trying to maximize what I got out of it. I don't have any times yet from this fella.
 
I'll bet it will be close. Post some smp times from it. That should give is a very good idea. Veeb0rg has an 2 x 4 dually with 8 cores at a slower speed that I never would have though would make the deadline, but does easily. Native Linux w/ BFS will hal a lot.


i believe veeborg is also running with triple channel memory though where as the ML350 G5 is running dual channel which will likely effect its ability to finish a bigadv WU even though it shows a time for SMP that is high enough to run bigadv WU's.
 
Those are running Core 2 Quad era CPUs with (presumably) relatively slow DDR2. I believe APOLLO had issues getting 2684s to finish on time at speeds below somewhere in the 2.8ghz range IIRC.

I'd imagine it will finish easier bigadv WUs on time but struggle with 2684s

Edit: If Veeb0rg is running triple channel DDR3 that means he's running Nehalem CPUs. That's apples and oranges unfortunately.
 
I have a dual quad core low energy dell 1950 that is just making the bigadv cutoff, I think it's avg comes out to 36kppd if I remember correctly

*edit* it's running CentOS 5 as well btw
 
Those systems shipped with E54** CPUs, which Apollo was running for a while.

does anyone know if he was running bigadv? I think it would be pretty damn close but im not sure as I skipped from E53** to E55**

Load up the client and lets she what shes got!
 
He was running bigadv, and I remember him saying 2684s were close to the deadline.
 
Those systems shipped with E54** CPUs, which Apollo was running for a while.

does anyone know if he was running bigadv? I think it would be pretty damn close but im not sure as I skipped from E53** to E55**

Load up the client and lets she what shes got!


he was running bigadv but it was a skull trail system which meant the processors were overclocked as well. but yeah 2.8Ghz was pretty much the limit for 2684's.
 
I just fired it up on bigadv and the first thing it grabbed was a 2684. Will report back tomorrow with times.
 
here are the times from a DL360 with e5420 2.5ghz procs

here's the avg from all wu's on this machine (total of 46 since switching it to linux bigadv)

P2684 = 47:57 (completes with 44% of time remaining)
P2685 = 35:31
P2686 = 33:40
P6901 = 33:35

*edit* btw if you want to verify your proc's model info more /proc/cpuinfo
 
[21:49:40]
[21:49:40] Project: 2684 (Run 1, Clone 21, Gen 62)
[21:49:40]
[21:49:40] Assembly optimizations on if available.
[21:49:40] Entering M.D.
:) G R O M A C S (-:

Groningen Machine for Chemical Simulation

:) VERSION 4.5.3 (-:

Written by Emile Apol, Rossen Apostolov, Herman J.C. Berendsen,
Aldert van Buuren, Pär Bjelkmar, Rudi van Drunen, Anton Feenstra,
Gerrit Groenhof, Peter Kasson, Per Larsson, Pieter Meulenhoff,
Teemu Murtola, Szilard Pall, Sander Pronk, Roland Schulz,
Michael Shirts, Alfons Sijbers, Peter Tieleman,

Berk Hess, David van der Spoel, and Erik Lindahl.

Copyright (c) 1991-2000, University of Groningen, The Netherlands.
Copyright (c) 2001-2010, The GROMACS development team at
Uppsala University & The Royal Institute of Technology, Sweden.
check out http://www.gromacs.org for more information.


:) Gromacs (-:

Reading file work/wudata_07.tpr, VERSION 4.0.99_development_20090605 (single precision)
[21:49:47] Mapping NT from 8 to 8
Note: tpx file_version 70, software version 73
Starting 8 threads
Making 1D domain decomposition 8 x 1 x 1
starting mdrun 'SINGLE VESICLE in water'
15750001 steps, 63000.0 ps (continuing from step 15500001, 62000.0 ps).
[21:49:54] Completed 0 out of 250000 steps (0%)
[23:02:49] Completed 2500 out of 250000 steps (1%)
[00:15:49] Completed 5000 out of 250000 steps (2%)
[01:28:50] Completed 7500 out of 250000 steps (3%)
[02:41:45] Completed 10000 out of 250000 steps (4%)
[03:47:12] - Autosending finished units... [May 13 03:47:12 UTC]
[03:47:12] Trying to send all finished work units
[03:47:12] + No unsent completed units remaining.
[03:47:12] - Autosend completed
[03:54:40] Completed 12500 out of 250000 steps (5%)
[05:07:35] Completed 15000 out of 250000 steps (6%)
[06:20:33] Completed 17500 out of 250000 steps (7%)
[07:33:29] Completed 20000 out of 250000 steps (8%)
[08:46:22] Completed 22500 out of 250000 steps (9%)
[09:47:12] - Autosending finished units... [May 13 09:47:12 UTC]
[09:47:12] Trying to send all finished work units
[09:47:12] + No unsent completed units remaining.
[09:47:12] - Autosend completed
[09:59:22] Completed 25000 out of 250000 steps (10%)
[11:12:18] Completed 27500 out of 250000 steps (11%)
[12:25:11] Completed 30000 out of 250000 steps (12%)

If my math is right it will be just a hair over five days. (If I recall that means I will get zero or at best base credit)

Well at least it is getting a proper CPU stability test running this unit.
 
I think 2684s have a little bit longer than the other WUs.

Tho that might be final deadline, not preferred.
 
I believe all bigadv have a preferred deadline of four days (for bonus points) and final deadline of six days.
 
I believe all bigadv have a preferred deadline of four days (for bonus points) and final deadline of six days.

At least I will get base credit and the science will count. About the time this unit is done this server will be heading out so it will probably only ever run this one unit.
 
Back
Top