Skip to content

.eqJoin().group() is slower than .map().group() #6063

Open

Description

@thelinuxlich

I'm using the query profiler and getting these results with a big table:

eqJoin().group()

4.95s round-trip time

630ms server time

130 shard accesses

map(with subquery).group()

517ms round-trip time

180ms server time

24 shard accesses

These are after reruns to warm the cache. Is this normal?

Activity

  1. danielmewes commented on Aug 23, 2016

    @danielmewes
    Member

    @thelinuxlich Is this on a cluster? How large is it? What are the sharding settings on the involved tables?

  2. added this to the subsequent milestone on Aug 23, 2016
  3. thelinuxlich commented on Aug 23, 2016

    @thelinuxlich
    Author

    It is a cluster with three machines, each table has 3 shards and 3 replicas

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions