You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As noted by @hammer and detailed in this post, Mongo has compression as of 2.8 via WiredTiger, so this may be less of a win, but still interesting to try.
Many field names today are taken directly from Spark's
JsonProtocol
and are quite verbose, e.g.TotalShuffleBytesRead
.Adding a layer that matches+replaces field-names before they go into Mongo would be nice, and may improve latency of
spree
server→client pubsub.The text was updated successfully, but these errors were encountered: