You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@impala.apache.org by "Gabriella Gyorgyevics (Code Review)" <ge...@cloudera.org> on 2025/12/16 10:29:52 UTC
[Impala-ASF-CR] IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
Gabriella Gyorgyevics has uploaded this change for review. ( http://gerrit.cloudera.org:8080/23770
Change subject: IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
......................................................................
IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
This patch adds benchmarks that compare Byte Stream Split encoding
to Dictionary and Delta encodings.
The performance is compared on the following types of datasets:
* Same data repeating (42, 42, 42,...)
* Filled up from a pool of data
* Sequential data (42, 43, 44, 45,...)
* Random data
The performance is compared for the following decoding modes:
* Reading one-by-one
* Reading in batch
* Reading with a stride
* Skipping over some data
These comparisons are made for both Dictionary and Delta.
Conclusions for Dictionary:
* The speed of dictionary decoding is very affected by
the type of the dataset it is decoding, while
BSS seems to be barely affected.
* Dictionary seems to be faster on every tested case,
except for reading one-by-one, where they are the
same speed
Conclusions for Delta:
* Delta also seems to be more affected by the type of
the dataset it is decoding.
* Similarly to Dictionary, Delta seems to be faster on
every test case except for reading one-by-one, where
Byte Stream Split is faster.
Change-Id: Ia4ad18d79c721d37147e5cd570e33caac8e10ac3
---
M be/src/benchmarks/parquet-byte-stream-split-decoder-benchmark.cc
1 file changed, 1,357 insertions(+), 55 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/Impala-ASF refs/changes/70/23770/3
--
To view, visit http://gerrit.cloudera.org:8080/23770
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: Ia4ad18d79c721d37147e5cd570e33caac8e10ac3
Gerrit-Change-Number: 23770
Gerrit-PatchSet: 3
Gerrit-Owner: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Impala Public Jenkins <im...@cloudera.com>
Gerrit-Reviewer: Zoltan Borok-Nagy <bo...@cloudera.com>
[Impala-ASF-CR] IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
Posted by "Impala Public Jenkins (Code Review)" <ge...@cloudera.org>.
Impala Public Jenkins has posted comments on this change. ( http://gerrit.cloudera.org:8080/23770 )
Change subject: IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
......................................................................
Patch Set 3: -Code-Review
Build Failed
https://jenkins.impala.io/job/gerrit-code-review-checks/21181/ : Initial code review checks failed. See linked job for details on the failure.
--
To view, visit http://gerrit.cloudera.org:8080/23770
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ia4ad18d79c721d37147e5cd570e33caac8e10ac3
Gerrit-Change-Number: 23770
Gerrit-PatchSet: 3
Gerrit-Owner: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Impala Public Jenkins <im...@cloudera.com>
Gerrit-Reviewer: Zoltan Borok-Nagy <bo...@cloudera.com>
Gerrit-Comment-Date: Tue, 16 Dec 2025 10:54:28 +0000
Gerrit-HasComments: No
[Impala-ASF-CR] IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
Posted by "Impala Public Jenkins (Code Review)" <ge...@cloudera.org>.
Impala Public Jenkins has posted comments on this change. ( http://gerrit.cloudera.org:8080/23770 )
Change subject: IMPALA-14387, IMPALA-14388: Compare the performance of Byte Stream Split encoding to Delta and Dictionary encodings
......................................................................
Patch Set 3: Verified-1
Build failed: https://jenkins.impala.io/job/gerrit-verify-dryrun/12799/
--
To view, visit http://gerrit.cloudera.org:8080/23770
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ia4ad18d79c721d37147e5cd570e33caac8e10ac3
Gerrit-Change-Number: 23770
Gerrit-PatchSet: 3
Gerrit-Owner: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Gabriella Gyorgyevics <gg...@cloudera.com>
Gerrit-Reviewer: Impala Public Jenkins <im...@cloudera.com>
Gerrit-Reviewer: Zoltan Borok-Nagy <bo...@cloudera.com>
Gerrit-Comment-Date: Tue, 16 Dec 2025 14:36:03 +0000
Gerrit-HasComments: No