QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: ko4 on October 12, 2021, 09:47

Title: Settings for 2x Intel Xeon 6226R
Post by: ko4 on October 12, 2021, 09:47

Hello!

I have a workstation based on 2 Intel Xeon 6226R (2x 16 Core, 64 threads).
OS is Win10.
Please tell me what are the optimal parallelization settings for solving 1, 2 and 4 tasks at the same time.

Thanks to all.

Title: Re: Settings for 2x Intel Xeon 6226R
Post by: mlee on October 15, 2021, 14:38

What kind of job do you run? Is it LCAO, Force field or PW? It depends on it. If you run the LCAO or PW, more mpi is better performance. (depending on the K-point) But if you run the force field, 1 node with multiple threads (ex, 8, 16) will be better performance. In fact, it depends on the hardware and conditions. You can test a few cases then you will be sure best one in your setting.

Title: Re: Settings for 2x Intel Xeon 6226R
Post by: ko4 on October 21, 2021, 09:07

Thanks for your reply.
Now I am working with LCAO.
As test have shown, the computation time for the same problem is reduced by 1.5 times compared to the i7-9700K. But with the simultaneous launch of 4 tasks (8 cores for each task) and when compared with the time spent on solving 4 tasks sequentially (in turn) on i7, the gain is already 3 times. When compared with two E5-2650 v2, we have the gain of more than 2 times. So, the processors really scale well.

Title: Re: Settings for 2x Intel Xeon 6226R
Post by: mlee on October 25, 2021, 11:30

Thanks for your updates.