Details
-
Improvement
-
Status: Closed
-
Major
-
Resolution: Abandoned
-
1.0.3
-
None
Description
Problem:
we experience some unexpected increase of data sent over the network for broadcasts with increasing number of slots per Taskmanager.
We provided a benchmark [1]. It not only increases the size of data sent over the network but also hurts performance as seen in the preliminary results below. In this results cloud-11 has 25 nodes and ibm-power has 8 nodes with scaling the number of slots per node from 1 - 16.
----------------------------------------------
suite | name | median_time |
==================================================
broadcast.cloud-11 | broadcast.01 | 8796 |
broadcast.cloud-11 | broadcast.02 | 14802 |
broadcast.cloud-11 | broadcast.04 | 30173 |
broadcast.cloud-11 | broadcast.08 | 56936 |
broadcast.cloud-11 | broadcast.16 | 117507 |
broadcast.ibm-power-1 | broadcast.01 | 6807 |
broadcast.ibm-power-1 | broadcast.02 | 8443 |
broadcast.ibm-power-1 | broadcast.04 | 11823 |
broadcast.ibm-power-1 | broadcast.08 | 21655 |
broadcast.ibm-power-1 | broadcast.16 | 37426 |
----------------------------------------------
After looking into the code base it, it seems that the data is de-serialized only once per TM, but the actual data is sent for all slots running the operator with broadcast vars and just gets discarded in case its already de-serialized.
We do not see a reason the data can't be shared among the slots of a TM and therefore just sent once.
[1] https://github.com/TU-Berlin-DIMA/flink-broadcast
This Jira will continue the discussion started here: https://mail-archives.apache.org/mod_mbox/flink-dev/201606.mbox/%3C1465386300767.94345@tu-berlin.de%3E