Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Szilárd Páll
Hi Anthony, Good choice, your admins should be able to help. Do emphasize when talking to them that you need your job to be placed as compact as possible, on the closest possible set of cores with the tightest possible affinity settings. Cheers, -- Szilárd On Thu, Aug 20, 2015 at 6:12 PM,

[gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Nash, Anthony
Hi all, I appear to have a very high load imbalance on some of my runs. Values starting from approx. 7% up to 31.8% with reported vol min/aver of around 0.6 (I haven¹t found one under half yet). When I look through the .log file at the start of the run I see: Initializing Domain Decomposition

Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Mark Abraham
Hi, In cases like this, it's good to describe what's in your simulation, and share the full .log file on a file-sharing service, so we can see both the things mdrun reports early and late. Mark On Thu, Aug 20, 2015 at 8:22 AM Nash, Anthony a.n...@ucl.ac.uk wrote: Hi all, I appear to have a

Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Nash, Anthony
Hi Mark, Many thanks for looking into this. One of the log files (the job hasn’t finished running) is here: https://www.dropbox.com/s/zwrro54yni2uxtn/umb_3_umb.log?dl=0 The system is a soluble collagenase in water with a collagen substrate and two zinc co-factors. There are 287562 atoms in the

Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Szilárd Páll
Hi, You're not pinning threads and it seems that you're running on a large SMP machine! Assuming that the 512 threads reported (line 91) is correct that's a 32 socket SMP machine, perhaps an SGI UV? In any case Xeon E5-4xxx is typically deployed in 4-8 socket installations, so your 8 threads will

Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Szilárd Páll
On Thu, Aug 20, 2015 at 5:52 PM, Szilárd Páll pall.szil...@gmail.com wrote: Hi, You're not pinning threads and it seems that you're running on a large SMP machine! Assuming that the 512 threads reported (line 91) is correct that's a 32 socket SMP machine, perhaps an SGI UV? In any case Xeon

Re: [gmx-users] High load imbalance: 31.8%

2015-08-20 Thread Nash, Anthony
Hi Szilárd Thanks for all of that advice. I’m going to have to take a lot of this up with the Cluster Service Staff. This is a new cluster service I won a grant for, thus not my usual platform which would typically yield an imbalance of somewhere around 0.8% to 2%. Thanks again Anthony On