Tim Armstrong has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/10256 )

Change subject: PREVIEW - IMPALA-6957: calc thread resource requirement in 
planner
......................................................................


Patch Set 4:

(1 comment)

I think we would want to allow rejection based on the per-host value (I'm not 
sure if that would use an upper bound or the actual max based on the schedule - 
maybe the first would be more robust), and also based on the global aggregate 
across fragments. One would protect any single daemon from resource exhaustive, 
the other would be more useful to protect a cluster from being overwhelmed.

http://gerrit.cloudera.org:8080/#/c/10256/4/testdata/workloads/functional-planner/queries/PlannerTest/spillable-buffer-sizing.test
File 
testdata/workloads/functional-planner/queries/PlannerTest/spillable-buffer-sizing.test:

http://gerrit.cloudera.org:8080/#/c/10256/4/testdata/workloads/functional-planner/queries/PlannerTest/spillable-buffer-sizing.test@6
PS4, Line 6: Max Per-Host Resource Reservation: Memory=18.97MB Threads=5
> what does the "Max" here refer to?
This is the maximum possible if all fragment instances were scheduled on a 
single backend.



--
To view, visit http://gerrit.cloudera.org:8080/10256
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I982837ef883457fa4d2adc3bdbdc727353469140
Gerrit-Change-Number: 10256
Gerrit-PatchSet: 4
Gerrit-Owner: Tim Armstrong <[email protected]>
Gerrit-Reviewer: Bikramjeet Vig <[email protected]>
Gerrit-Reviewer: Dan Hecht <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-Comment-Date: Mon, 07 May 2018 21:17:45 +0000
Gerrit-HasComments: Yes

Reply via email to