[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zhaojing Yu updated HUDI-2928:
--
Fix Version/s: 0.13.0
(was: 0.12.1)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.13.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Sprint: (was: 2022/09/05)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.1
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
sivabalan narayanan updated HUDI-2928:
--
Sprint: 2022/09/05 (was: 2022/09/19)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.1
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
sivabalan narayanan updated HUDI-2928:
--
Story Points: 2
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.1
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Sprint: 2022/09/19 (was: 2022/08/22)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.1
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sagar Sumit updated HUDI-2928:
--
Fix Version/s: 0.12.1
(was: 0.12.0)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.1
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Sprint: Hudi-Sprint-Jan-10, 2022/08/08 (was: Hudi-Sprint-Jan-10)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
sivabalan narayanan updated HUDI-2928:
--
Sprint: Hudi-Sprint-Jan-10 (was: Hudi-Sprint-Jan-10, 2022/05/02)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Priority: Critical (was: Blocker)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Critical
> Labels: pull-request-available
> Fix For: 0.12.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Sprint: Hudi-Sprint-Jan-10, 2022/05/02 (was: Hudi-Sprint-Jan-10)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.12.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Component/s: performance
storage-management
Epic Link: HUDI-3249
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Issue Type: Improvement (was: Task)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Fix Version/s: 0.12.0
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Improvement
> Components: performance, storage-management
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.12.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Raymond Xu updated HUDI-2928:
-
Fix Version/s: (was: 0.11.0)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Sprint: Hudi-Sprint-Jan-10 (was: Hudi-Sprint-Jan-10, Hudi-Sprint-Jan-18)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinoth Chandar updated HUDI-2928:
-
Sprint: Hudi-Sprint-Jan-10, Hudi-Sprint-Jan-18 (was: Hudi-Sprint-Jan-10)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinoth Chandar updated HUDI-2928:
-
Status: Open (was: In Progress)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Status: In Progress (was: Open)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Sprint: Hudi-Sprint-Jan-10
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Status: Open (was: In Progress)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Status: In Progress (was: Open)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinoth Chandar updated HUDI-2928:
-
Status: Open (was: Patch Available)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinoth Chandar updated HUDI-2928:
-
Priority: Blocker (was: Major)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Blocker
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Attachment: Screen Shot 2021-12-06 at 11.49.05 AM.png
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png, Screen Shot
> 2021-12-06 at 11.49.05 AM.png, image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Status: Patch Available (was: In Progress)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png,
> image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Description:
Currently, having Gzip as a default we prioritize Compression/Storage cost at
the expense of
* Compute (on the {+}write-path{+}): about *30%* of Compute burned during
bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
* Compute (on the {+}read-path{+}), as well as queries Latencies: queries
scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put is
*3-4x* less than Snappy, Zstd,
[EX|https://stackoverflow.com/a/56410326/3520840])
P.S Spark switched its default compression algorithm to Snappy [a while
ago|https://github.com/apache/spark/pull/12256].
*EDIT*
We should actually evaluate putting in
[zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
instead of Snappy. It has compression ratios comparable to Gzip, while
bringing in much better performance:
!image-2021-12-03-13-13-02-892.png!
[https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
was:
Currently, having Gzip as a default we prioritize Compression/Storage cost at
the expense of
* Compute (on the {+}write-path{+}): about *30%* of Compute burned during
bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
* Compute (on the {+}read-path{+}), as well as queries Latencies: queries
scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put is
*3-4x* less than Snappy, Zstd,
[EX|https://stackoverflow.com/a/56410326/3520840])
P.S Spark switched its default compression algorithm to Snappy [a while
ago|https://github.com/apache/spark/pull/12256].
*EDIT*
We should actually evaluate putting in
[zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
instead of Snappy. It has compression ratios comparable to Gzip, while
bringing in much better performance:
!image-2021-12-03-13-13-02-892.png!
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png,
> image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
> [https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ASF GitHub Bot updated HUDI-2928:
-
Labels: pull-request-available (was: )
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Labels: pull-request-available
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png,
> image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Fix Version/s: 0.11.0
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Fix For: 0.11.0
>
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png,
> image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
[jira] [Updated] (HUDI-2928) Evaluate rebasing Hudi's default compression from Gzip to Zstd
[
https://issues.apache.org/jira/browse/HUDI-2928?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Kudinkin updated HUDI-2928:
--
Description:
Currently, having Gzip as a default we prioritize Compression/Storage cost at
the expense of
* Compute (on the {+}write-path{+}): about *30%* of Compute burned during
bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
* Compute (on the {+}read-path{+}), as well as queries Latencies: queries
scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put is
*3-4x* less than Snappy, Zstd,
[EX|https://stackoverflow.com/a/56410326/3520840])
P.S Spark switched its default compression algorithm to Snappy [a while
ago|https://github.com/apache/spark/pull/12256].
*EDIT*
We should actually evaluate putting in
[zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
instead of Snappy. It has compression ratios comparable to Gzip, while
bringing in much better performance:
!image-2021-12-03-13-13-02-892.png!
was:
Currently, having Gzip as a default we prioritize Compression/Storage cost at
the expense of
* Compute (on the {+}write-path{+}): about *30%* of Compute burned during
bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
* Compute (on the {+}read-path{+}), as well as queries Latencies: queries
scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put is
*3-4x* less than Snappy's, [EX|https://stackoverflow.com/a/56410326/3520840])
P.S Spark switched its default compression algorithm to Snappy [a while
ago|https://github.com/apache/spark/pull/12256].
*EDIT*
We should actually evaluate putting in
[zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
instead of Snappy. It has compression ratios comparable to Gzip, while
bringing in much better performance:
!image-2021-12-03-13-13-02-892.png!
Summary: Evaluate rebasing Hudi's default compression from Gzip to Zstd
(was: Evaluate rebasing Hudi's default compression from Gzip to Snappy)
> Evaluate rebasing Hudi's default compression from Gzip to Zstd
> --
>
> Key: HUDI-2928
> URL: https://issues.apache.org/jira/browse/HUDI-2928
> Project: Apache Hudi
> Issue Type: Task
>Reporter: Alexey Kudinkin
>Assignee: Alexey Kudinkin
>Priority: Major
> Attachments: Screen Shot 2021-12-03 at 12.36.13 PM.png,
> image-2021-12-03-13-13-02-892.png
>
>
> Currently, having Gzip as a default we prioritize Compression/Storage cost at
> the expense of
> * Compute (on the {+}write-path{+}): about *30%* of Compute burned during
> bulk-insert in local benchmarks on Amazon Reviews dataset is Gzip (see below)
> * Compute (on the {+}read-path{+}), as well as queries Latencies: queries
> scanning large datasets are likely to be compression-/CPU-bound (Gzip t/put
> is *3-4x* less than Snappy, Zstd,
> [EX|https://stackoverflow.com/a/56410326/3520840])
> P.S Spark switched its default compression algorithm to Snappy [a while
> ago|https://github.com/apache/spark/pull/12256].
>
> *EDIT*
> We should actually evaluate putting in
> [zstd|https://engineering.fb.com/2016/08/31/core-data/smaller-and-faster-data-compression-with-zstandard/]
> instead of Snappy. It has compression ratios comparable to Gzip, while
> bringing in much better performance:
> !image-2021-12-03-13-13-02-892.png!
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
