Speed up shard rebalance

I couldn't find anything recent on this topic.

I have 4 data nodes, excluded one from routing

PUT _cluster/settings
  "transient" :{
      "cluster.routing.allocation.exclude._ip" : ""

and added 5th one to compensate for missing storage, while I replacing excluded node.

Now I have 5 servers, 8CPU each, 64 GB of RAM and 10GBps network interfaces (seating in AWS) barely loaded:
I/O wait is about 2-5 % per core
CPU is only 100% (out of 800)
Network and disks are doing nothing

200+ shards are relocating, but I don't see much of a progress there. Any thoughts?

Thank you

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.