[Beowulf] How to debug error with Open MPI 3 / Mellanox / Red Hat?

Faraz Hussain info at feacluster.com
Thu May 2 11:09:42 PDT 2019


Thanks Benson, these are very useful links. I browsed the Mellanox  
guide and it seems the target audience is experts in networking. I  
wish there was some quick start guide or Infiniband for dummies book :-)



Quoting Benson Muite <benson_muite at emailplus.org>:

> Hi Faraz,
>
> Mellanox manuals can be found at:
>
> https://docs.mellanox.com/
>
> Example setup instructions (not sure if correct for you as do not  
> have exact details on your hardware):
>
> https://www.mellanox.com/related-docs/prod_software/Mellanox_OFED_Linux_User_Manual_v4_3.pdf
>
> Maybe also helpful (students who have participated in cluster  
> competitions are usually quite good at setting these up):
>
> https://www.slothparadise.com/setting-infiniband-centos-6-7/
>
> If you will be primarily running finite element software,  time  
> investment now for understanding performance analysis can pay off in  
> future - Disclosure I have an interest in seeing more performance  
> tests on your hardware.
>




More information about the Beowulf mailing list