|
[Sponsors] |
March 28, 2008, 07:19 |
particle tracking
|
#1 |
Guest
Posts: n/a
|
Hello,
I am solving combusting flow on a linux cluster and I notice the DPM update is taking quite a while even on 8 compute nodes. I see the parallel particle tracking concurrency is only around 4%. Is there any way to improve the concurrency? Thanks in advance RE13 |
|
March 28, 2008, 14:28 |
Re: particle tracking
|
#2 |
Guest
Posts: n/a
|
I am not aware of means for improving particle tracking concurrency. I am usually solving with unsteady DPM on 8 CPU and I see anywhere from 10% to 40% concurrency. In my case most of the CPU time (99.9%) is spent for DPM iterations. My domain is 2D axi and usually contains 500,000 to 1,000,000 particles.
Remus |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
CFX particle tracking | Mehmet SORGUN | CFX | 0 | February 4, 2009 13:24 |
particle injection and particle tracking | yair | FLUENT | 0 | July 15, 2008 08:27 |
Particle Tracking | ED | FLUENT | 0 | August 7, 2007 01:29 |
Particle tracking | nicostage | CFX | 1 | May 4, 2006 09:08 |
about particle tracking! | shirley | FLUENT | 0 | June 25, 2003 00:34 |