Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

This page describes issues we have had with the cluster (most recent ones first) and how we have dealt with them.

  • Strange bandwidth issues:  ib_send_bw tests on second rack (rc41-rc80 with ConnectX-3 nics and 56Gbps SwitchX switches) only show ~1650MB/s between nodes.
    1. Bandwidths are apparently fine (~3200MB/s) between rc01-rc40, which are on th old switches (though they route through the new switches)
    2. Strangely, bandwidths between nodes in first rack and second rack are fine (~3200MB/s)
      1. So it appears to be a combination of new nics talking to new nics!?

...