phoenix-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From alex kamil <>
Subject Re: Phoenix and SQL In clauses
Date Wed, 28 May 2014 17:53:27 GMT
we've tried something similar, on a table with a few hundreds of VARCHAR
columns getting   >30k items in IN clause was starting to exceed 60 sec, if
I remember correctly,
it basically becomes an IO bottleneck getting the huge resultsets back to
the client

this was a workaround we used before JOINs became available in phoenix -
pulled a subset of data to the client  using a batch of M queries with N
elements each, and did a join on client side in HSQLDB, which was messy and
with phoenix 3.0 we switched to using joins on server side and avoid this
scenario completely


On Wed, May 28, 2014 at 1:27 PM, Dmitry Goldenberg

> Hi,
> I was wondering that the maximum value of elements in an IN clause is, in
> Phoenix.  Also, from the performance standpoint, if I were to batch up the
> items into M statements of N elements per In clause and execute multiple
> statements to avoid ever hitting the max, what may be an optimal value for
> N, considering that I only have a few relatively small column values (int,
> varchar, timestamp) per statement?
> Thanks.
> --
> You received this message because you are subscribed to the Google Groups
> "Phoenix HBase User" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to
> For more options, visit

View raw message