asterixdb-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xikui Wang (Code Review)" <do-not-re...@asterixdb.incubator.apache.org>
Subject Change in asterixdb[master]: [ASTERIXDB-2152][FUN][COMP] Enable specifying computation lo...
Date Tue, 07 Nov 2017 18:40:33 GMT
Xikui Wang has posted comments on this change.

Change subject: [ASTERIXDB-2152][FUN][COMP] Enable specifying computation location
......................................................................


Patch Set 12:

(4 comments)

Several replies. Thanks Till!

https://asterix-gerrit.ics.uci.edu/#/c/2114/12/asterixdb/asterix-app/src/test/resources/runtimets/queries_sqlpp/feeds/feed-with-external-function/feed-with-external-function.3.update.sqlpp
File asterixdb/asterix-app/src/test/resources/runtimets/queries_sqlpp/feeds/feed-with-external-function/feed-with-external-function.3.update.sqlpp:

PS12, Line 25: "7"
> Is there a way to validate, that the job runs using this degree of parallel
Yes, there is. The computation locations are decided at the compilation phase, so we could
validate that before we run. My question is do we want to validate that? The parallelism option
is designed as a "hint", i.e., not forced. For example, if the cluster has 6 cores in total,
the query will only be running on 6 threads in parallel.


https://asterix-gerrit.ics.uci.edu/#/c/2114/12/asterixdb/asterix-common/src/main/java/org/apache/asterix/common/config/CompilerProperties.java
File asterixdb/asterix-common/src/main/java/org/apache/asterix/common/config/CompilerProperties.java:

Line 134:     
> MAJOR SonarQube violation:
Will fix in next patch.


https://asterix-gerrit.ics.uci.edu/#/c/2114/12/asterixdb/asterix-common/src/main/resources/asx_errormsg/en.properties
File asterixdb/asterix-common/src/main/resources/asx_errormsg/en.properties:

PS12, Line 121: Invalid computation location
> Could this contain the invalid location(s)?
Since I added an additional compiler option for specifying the computing location, it's possible
that the node id that user provided is not valid.  During the query compilation, the specified
computation locations are validated and the exception will be thrown if the validation fails.


https://asterix-gerrit.ics.uci.edu/#/c/2114/12/hyracks-fullstack/algebricks/algebricks-core/src/main/java/org/apache/hyracks/algebricks/core/algebra/operators/physical/AssignPOperator.java
File hyracks-fullstack/algebricks/algebricks-core/src/main/java/org/apache/hyracks/algebricks/core/algebra/operators/physical/AssignPOperator.java:

PS12, Line 118: setLocationConstraint
> Why is is ok to replace the cardinality constraint with a location constrai
If the node replacement happens before the query compilation, the query with the only node
id will be invalidated by the computation location validation in APIFrameWork. If the node
replacement happens after the compilation but before the execution, the query will be invalidated.
But this is a problem for other queries involve LocationContraints....


-- 
To view, visit https://asterix-gerrit.ics.uci.edu/2114
To unsubscribe, visit https://asterix-gerrit.ics.uci.edu/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Id7eed5dac03c2f260507e16cf687162d65787bd1
Gerrit-PatchSet: 12
Gerrit-Project: asterixdb
Gerrit-Branch: master
Gerrit-Owner: Xikui Wang <xkkwww@gmail.com>
Gerrit-Reviewer: Anon. E. Moose #1000171
Gerrit-Reviewer: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Gerrit-Reviewer: Till Westmann <tillw@apache.org>
Gerrit-Reviewer: Xikui Wang <xkkwww@gmail.com>
Gerrit-HasComments: Yes

Mime
View raw message