[Beowulf] If I were specifying a new custer...
deadline at eadline.org
Thu Oct 11 12:08:33 PDT 2018
Over the last several months I have been reading about:
2) Intel Fab issues
3) Supermicro MB issues
I started thinking, if I were going to specify a
single rack cluster, what would I use?
I'm assuming a general HPC workload (not deep learning or
analytics) I need to choose Xeon/Epyc, IB/Omni,
Lustre/Ceph/BeGFS, should all nodes have GPU?
I'm interested what members of this list think?
More information about the Beowulf