[Beowulf] Best case performance of HPL on EPYC 7742 processor ...

[Joshua Mora]

Reach out AMD,
they have specific instructions (including BIOS/OS settings) and even binaries
on how to get the best performance.
Dont go try and error as is very time consuming.
BLIS has also multiple parameters as it has nested loops, so you could also
have to try multiple configurations to get the optimal performance.
Just reach to them.

Joshua

------ Original Message ------
Received: 04:30 PM CDT, 08/14/2020
From: Richard Walsh <rbwcnslt at gmail.com>
To: Beowulf List <beowulf at beowulf.org>
Subject: [Beowulf] Best case performance of HPL on EPYC 7742 processor ...

> All,
> 
> What have people achieved on this SKU on a single-node using the stock
> HPL 2.3 source... ??
> 
> I have seen a variety of performance claims even as high as 90% of its
> nominal
> per node peak of 4.608 TFLOPs.  I can now get above 80% of peak, but not
> higher.
> I have heard that to get higher values special BIOS settings are required,
> including
> the turning off SMT which allows the chip to turbo higher.  Remember this
> is not the
> 7542 processor with 32 cores per chip and the same bandwidth per socket as
> the
> 7742 which can turbo to over 100% of nominal peak for HPL.
> 
> If people have gotten higher single node numbers ... what is your recipe
> ... ??
> 
> I am particularly interested in BIOS settings, and maybe surprise settings
> in the HPL.dat file.  Do higher performing runs require using close to the
> maximum memory on the node ... ??  As this is single-node, I would not
> expect choice of MPI to make a difference
> 
> To get to 80% with SMT on in the BIOS, I am building with an older Intel
> compiler and MKL that still recognizes the MKL_DEBUG_CPU_TYPE=5.
> Running so that the number of MPI ranks run on the node matches the
> number of CCXs seems ot give the best numbers.
> 
> Following the tuning instructions from AMD for using BLIS and GCC for
> the build does not get me there.
> 
> Thanks,
> 
> Richard Walsh
> 

> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
> To change your subscription (digest mode or unsubscribe) visit
https://beowulf.org/cgi-bin/mailman/listinfo/beowulf