| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
platinumjsi
Cruncher Joined: Aug 12, 2010 Post Count: 37 Status: Offline Project Badges:
|
Quick one, do these unit times seem right?
----------------------------------------PC 1 = 5930k CPU @3.7GHz, HT enabled MCM 7.35 taking around 8 hours per unit This machine also runs GPU units but I have it set to only use 9 cores to leave some free for the GPU, the CPU is never maxed out, sits around 90% used PC2 = 4590T CPU @3ghz (Turbos to this all day never drops) no HT MCM 7.35 on this chip, taking around 3:30 - 4 hours per unit No GPU on this, all cores on WCG PC 3 (linux) 3770k running stock speeds HT enabled, MCM 7.35 taking around 3 hours per unit No GPU on this, running all 8 threads. So looking at this the fastest CPU I have seems to be very slow at the WU's, any thoughts? task manager shows the CPU load at 90% so its defiantly hammering the CPU? ![]() |
||
|
|
platinumjsi
Cruncher Joined: Aug 12, 2010 Post Count: 37 Status: Offline Project Badges:
|
Just done a manual update on the client and things seem to be moving faster, estimates of 3 hours on the latest blocks now, is it possible for some reason for things to slow over time??
----------------------------------------![]() |
||
|
|
ashrader330
Advanced Cruncher Joined: Jan 6, 2008 Post Count: 97 Status: Offline Project Badges:
|
MCM work units have a very high variability in run time. I have a i7 3770 and I have seen work units vary from 1 hour to 18 hours on that machine. That is just the way the work for this project is. It is completely normal.
----------------------------------------EDIT: Take two of the work units I have currently running on the machine I am using right now. Work unit MCM1_0012865_7372_0 has taken over 7 hours to reach 69% while work unit MCM1_0012867_4646_0 has taken a little more than 1.5 hours to reach 50%. ![]() Run time: 4.2y HPF2, 6.9y FAAH, 7.9y HFCC, 20.8y HCC, 26.0y CEP2, 26.0y MCM, 2.1y UGM, 2.0y OET WU: 4.8k HPF2, 12.3k FAAH, 12.4k HFCC, 135k HCC, 34.3k CEP2, 43.8k MCM, 4.2k UGM, 19.7k OET [Edit 1 times, last edit by ashrader330 at Apr 2, 2015 8:01:44 PM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7844 Status: Offline Project Badges:
|
Ashrader330 is correct. There is a fair amount of variability in the WU length. If you are only running MCM1 the way to compare the machines is to divide the the points granted by the cpu time. This will give you figures in points per hour. On average, your faster machines should be getting more points per hour than your slower machines. If you track this for several days, you should see the differences in the machines. Good luck
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges:
|
Quick one, do these unit times seem right? PC 1 = 5930k CPU @3.7GHz, HT enabled MCM 7.35 taking around 8 hours per unit This machine also runs GPU units but I have it set to only use 9 cores to leave some free for the GPU, the CPU is never maxed out, sits around 90% used you can crunch on 10cores...on the BOINC task list you can see how much CPU is GPU client taking...probably just 5-25% of core! ;) |
||
|
|
platinumjsi
Cruncher Joined: Aug 12, 2010 Post Count: 37 Status: Offline Project Badges:
|
Cheers, nah 9 is about the most I can get out of it, leaves enough headroom for the GPU and for background processes :)
----------------------------------------![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Whilst the inter-task variability may still be great [?], something has changed in the task builds process or the compounds being crunched, for since early March the daily -average- run-times for MCM1 are fairly stable, near flattish. See http://bit.ly/WCGART , grey curve.
|
||
|
|
|