Pooja Nilangekar has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11517 )

Change subject: [WIP] IMPALA-6932: Speed up scans for sequence datasets with 
many files
......................................................................


Patch Set 3:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/exec/hdfs-scan-node-base.h
File be/src/exec/hdfs-scan-node-base.h:

http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/exec/hdfs-scan-node-base.h@262
PS2, Line 262:       EnqueueLocation enqueue_location = EnqueueLocation::TAIL) 
WARN_UNUSED_RESULT;
> I think this is a reasonable place to use a default argument value but I'm
Agreed.
Actually, thinking about it made me realize that this is probably a bad idea. 
If we were to add another scanner for a new file format, we'd definitely use 
some variant of the AddDiskIoRanges() function. At that point it would make 
sense for the caller to think/reason about the queueing order.  (Hopefully 
avoiding behavior like this issue.)
I tried getting rid of the default argument but it caused issues with the 
HdfsLzoTextScanner::IssueFileRanges() in Impala-lzo. I have changed all the 
invocations to pass in the EnqueueLocation and will get rid of the default 
arguments once Impala-lzo is modified. Does that make sense? Or should I let 
the default parameters remain as is?


http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/exec/hdfs-scan-node.h
File be/src/exec/hdfs-scan-node.h:

http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/exec/hdfs-scan-node.h@90
PS2, Line 90:       int num_files_queued, EnqueueLocation enqueue_location =
> weird line wrapping. Ok to leave if clang-format did this.
Yes, it was because of clang-format.


http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/runtime/io/request-context.h
File be/src/runtime/io/request-context.h:

http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/runtime/io/request-context.h@28
PS2, Line 28: L
> extra space
Done


http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/runtime/io/request-context.cc
File be/src/runtime/io/request-context.cc:

http://gerrit.cloudera.org:8080/#/c/11517/2/be/src/runtime/io/request-context.cc@358
PS2, Line 358:         DCHECK_ENUM_EQ(schedule_mode, 
ScheduleMode::UPON_GETNEXT_HEAD);
> Add a DCHECK_ENUM_EQ to assert that it's UPON_GETNEXT_HEAD so that the assu
Done



--
To view, visit http://gerrit.cloudera.org:8080/11517
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I211e2511ea3bb5edea29f1bd63e6b1fa4c4b1965
Gerrit-Change-Number: 11517
Gerrit-PatchSet: 3
Gerrit-Owner: Pooja Nilangekar <[email protected]>
Gerrit-Reviewer: Bikramjeet Vig <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Pooja Nilangekar <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-Comment-Date: Mon, 22 Oct 2018 20:58:37 +0000
Gerrit-HasComments: Yes

Reply via email to