|
[Sponsors] |
January 1, 2000, 07:46 |
Parallelization efficiency..
|
#1 |
Guest
Posts: n/a
|
Hi all,
I am doing explicit unstructured Finite Volume CFD on a 24 node message passing parallel computer. I am getting efficiencies of more than 100% (around 115%) when using 2 & 4 nodes. The problem size is relatively small (< 7 MB per node) comapared to the available memory (30 MB per node). Can anyone attribute reasons towards this?. Also, the efficiency is always maximum when i'm using 4 nodes. I have run codes for 4 different levels of grid and 'am consistently getting the same numbers. Note: eff = time on 1 proc / (n * time on n proc) time : time per iteration. |
|
January 4, 2000, 12:20 |
Re: Parallelization efficiency..
|
#2 |
Guest
Posts: n/a
|
Efficiencies greater than 100% are quite common for codes which are well suited to parallelisation. (This is termed "super-linear speed-up"). The reason is supposedly to do with increases in the efficiency of the cache use (at least on distributed memory machines).
As the number of processors increases the efficiency will then decrease because of the extra message passing so you should expect a maximum at some point. Hope this helps a little. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
MPI Aborts when using reactingFoam with cyclic BC in parallelization | jose_rodrig | OpenFOAM | 1 | February 14, 2011 04:05 |
Efficiency, thrust/drag with regard to hovering | quarkz | Main CFD Forum | 0 | April 21, 2010 01:51 |
Droplet Collision Efficiency Calculation using DPM | Virk | FLUENT | 0 | January 22, 2010 10:37 |
Parelleling Efficiency | kassiotis | OpenFOAM | 0 | June 19, 2009 15:12 |
Efficiency for hovering flight? | Zonexo | Main CFD Forum | 1 | April 14, 2008 01:49 |