Slow recovery speeds


(Avaneesh) #1

Hi all!
ES v2.4.0
I am using a AWS i3.4xlarge instance (NVMe SSDs) which is IO optimized and has high Network throughput
Single node cluster- 16 cores, 120G memory, 10gbps network link
I try to restore a snapshot from S3 repository (size: 320G, 3 shards) to this node
I check the values in iotop and iftop, and they confirm that the recovery is restricted by the default 40mb max_bytes_per_sec limit.

Then I add these settings to my cluster:

curl -XPUT 10.5.2.30:9200/_cluster/settings -d '{
"persistent" : {
"cluster.routing.allocation.node_concurrent_recoveries": 20,
"indices.recovery.max_bytes_per_sec": "100mb",
"indices.recovery.concurrent_streams": 5,
"indices.recovery.concurrent_small_file_streams": 5,
"indices.recovery.translog_ops": 5000,
"cluster.routing.allocation.node_initial_primaries_recoveries": 30}
}'

I restart the cluster and try to recover the snapshot.
But I still get the exact same speeds - 40MBps write.

Where am I missing?


(system) #2

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