[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benjamin Lerer updated CASSANDRA-15511:
---
Fix Version/s: 4.0.5
4.1-alpha
(was: 4.x)
(was: 4.0.x)
(was: 4.1.x)
Source Control Link:
https://github.com/apache/cassandra/commit/c378874a9fa123891d1d75177d99dba5c4d18f9b
Resolution: Fixed
Status: Resolved (was: Ready to Commit)
Committed into cassandra-4.0 at c378874a9fa123891d1d75177d99dba5c4d18f9b and
merged into cassandra-4.1 and trunk
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.5, 4.1-alpha
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benjamin Lerer updated CASSANDRA-15511:
---
Status: Ready to Commit (was: Review In Progress)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.1.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Caleb Rackliffe updated CASSANDRA-15511:
Fix Version/s: 4.1.x
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.1.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Branimir Lambov updated CASSANDRA-15511:
Reviewers: Benjamin Lerer, Branimir Lambov (was: Benjamin Lerer)
Status: Review In Progress (was: Patch Available)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benjamin Lerer updated CASSANDRA-15511:
---
Test and Documentation Plan: The changes are covered by exisitng unit tests
(was: AtomicBTreePartitionUpdateBench included. Some unit tests to come.)
Status: Patch Available (was: In Progress)
The PR for the 4.0 branch can be found
[here|https://github.com/apache/cassandra/pull/1646] and the CI results can be
found
[here|https://app.circleci.com/pipelines/github/blerer/cassandra/300/workflows/6950b3c5-168b-43c4-a9d8-a16bdcace345]
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benjamin Lerer updated CASSANDRA-15511:
---
Status: In Progress (was: Patch Available)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benjamin Lerer updated CASSANDRA-15511:
---
Reviewers: Benjamin Lerer
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Michael Semb Wever updated CASSANDRA-15511:
---
Fix Version/s: 4.0.x
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Attachment: atomicbtreepartition.xlsx.zip
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Attachment: (was: atomicbtreepartition.xlsx.zip)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Fix Version/s: 4.x
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Fix For: 4.x
>
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Description:
This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
produced by a number of common operations, by employing {{transformAndFilter}},
{{transform}} and {{FastBuilder}}
* {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
{{BTree.transform}}, so no special builders are necessary;
** {{Rows.copy}} removed
* {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
reconciler
** Zero-allocations if result of merge is same as {{existing}}
** Fewer comparisons
* {{ColumnData}} reconciler implemented in same manner
** {{Cells.reconcileComplex}} is retired
** {{ComplexColumnData}} reconciliation now
*** Garbage-free if the merge has no effect
*** Always fewer allocations
*** Fewer comparisons
* {{FastBuilder}} employed widely:
** {{ClusteringIndexNamesFilter}} deserialization
** {{Columns}} deserialization
** {{PartitionUpdate}} deserialization
** {{AbstractBTreePartition}} construction
** Misc others
The upshot of this work when combined with the proposed patch for
CASSANDRA-15367 has a dramatic impact on operations over collection types -
under contention, as much as 100x improved throughput, and hundreds of
megabytes of reduced allocations. For all operations, allocations under
contention and no contention are significantly reduced and throughput improved.
was:
This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
produced by a number of common operations, by employing {{transformAndFilter}},
{{transform}} and {{FastBuilder}}
* {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
{{BTree.transform}}, so no special builders are necessary;
** {{Rows.copy}} removed
* {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
reconciler
** Zero-allocations if result of merge is same as {{existing}}
** Fewer comparisons
* {{ColumnData}} reconciler implemented in same manner
** {{Cells.reconcileComplex}} is retired
** {{ComplexColumnData}} reconciliation now
*** Garbage-free if the merge has no effect
*** Always fewer allocations
*** Fewer comparisons
* {{FastBuilder}} employed widely:
** {{ClusteringIndexNamesFilter}} deserialization
** {{Columns}} deserialization
** {{PartitionUpdate}} deserialization
** {{AbstractBTreePartition}} construction
** Misc others
The upshot of this work when combined with the proposed patch for
CASSANDRA-15367 has a dramatic impact on operations over collection types -
under contention, as much as 100x improved throughput, and hundreds of
megabytes of reduced allocations. For all operations, allocations under
contention and no contention are significantly reduced and throughput improved.
I am still awaiting the final results of the comprehensive performance
comparison for {{AtomicBTreePartition}}, after which I will establish what
impact there might be on compaction and normal reads, both of which should be
affected by this patch.
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations,
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Attachment: atomicbtreepartition.xlsx.zip
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Attachments: atomicbtreepartition.ods, atomicbtreepartition.xlsx.zip,
> perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Attachment: perfsh.tar.gz
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Attachments: atomicbtreepartition.ods, perfsh.tar.gz
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Attachment: atomicbtreepartition.ods
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
> Attachments: atomicbtreepartition.ods
>
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Test and Documentation Plan: some unit tests to come
Status: Patch Available (was: Open)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Test and Documentation Plan: AtomicBTreePartitionUpdateBench included. Some
unit tests to come. (was: some unit tests to come)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Description:
This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
produced by a number of common operations, by employing {{transformAndFilter}},
{{transform}} and {{FastBuilder}}
* {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
{{BTree.transform}}, so no special builders are necessary;
** {{Rows.copy}} removed
* {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
reconciler
** Zero-allocations if result of merge is same as {{existing}}
** Fewer comparisons
* {{ColumnData}} reconciler implemented in same manner
** {{Cells.reconcileComplex}} is retired
** {{ComplexColumnData}} reconciliation now
*** Garbage-free if the merge has no effect
*** Always fewer allocations
*** Fewer comparisons
* {{FastBuilder}} employed widely:
** {{ClusteringIndexNamesFilter}} deserialization
** {{Columns}} deserialization
** {{PartitionUpdate}} deserialization
** {{AbstractBTreePartition}} construction
** Misc others
The upshot of this work when combined with the proposed patch for
CASSANDRA-15367 has a dramatic impact on operations over collection types -
under contention, as much as 100x improved throughput, and hundreds of
megabytes of reduced allocations. For all operations, allocations under
contention and no contention are significantly reduced and throughput improved.
I am still awaiting the final results of the comprehensive performance
comparison for {{AtomicBTreePartition}}, after which I will establish what
impact there might be on compaction and normal reads, both of which should be
affected by this patch.
was:
This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
produced by a number of common operations, by employing {{transformAndFilter}},
{{transform}} and {{FastBuilder}}
{{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
{{BTree.transform}}, so no special builders are necessary; {{Rows.copy}} removed
{{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
reconciler
Zero-allocations if result of merge is same as {{existing}}
Fewer comparisons
{{ColumnData}} reconciler implemented in same manner
{{Cells.reconcileComplex}} is retired
{{ComplexColumnData}} reconciliation now
Garbage-free if the merge has no effect
Always fewer allocations
Fewer comparisons
{{FastBuilder}} employed widely:
{{ClusteringIndexNamesFilter}} deserialization
{{Columns}} deserialization
{{PartitionUpdate}} deserialization
{{AbstractBTreePartition}} construction
Misc others
The upshot of this work when combined with the proposed patch for
CASSANDRA-15367 has a dramatic impact on operations over collection types -
under contention, as much as 100x improved throughput, and hundreds of
megabytes of reduced allocations. For all operations, allocations under
contention and no contention are significantly reduced and throughput improved.
I am still awaiting the final results of the comprehensive performance
comparison for {{AtomicBTreePartition}}, after which I will establish what
impact there might be on compaction and normal reads, both of which should be
affected by this patch.
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> * {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary;
> ** {{Rows.copy}} removed
> * {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> ** Zero-allocations if result of merge is same as {{existing}}
> ** Fewer comparisons
> * {{ColumnData}} reconciler implemented in same manner
> ** {{Cells.reconcileComplex}} is retired
> ** {{ComplexColumnData}} reconciliation now
> *** Garbage-free if the merge has no effect
> *** Always fewer allocations
> *** Fewer comparisons
> * {{FastBuilder}} employed widely:
> ** {{ClusteringIndexNamesFilter}} deserialization
> ** {{Columns}} deserialization
> ** {{PartitionUpdate}} deserialization
> ** {{AbstractBTreePartition}} construction
> ** Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as
[jira] [Updated] (CASSANDRA-15511) Utilising BTree Improvements
[
https://issues.apache.org/jira/browse/CASSANDRA-15511?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Benedict Elliott Smith updated CASSANDRA-15511:
---
Change Category: Performance
Complexity: Normal
Assignee: Benedict Elliott Smith
Status: Open (was: Triage Needed)
> Utilising BTree Improvements
>
>
> Key: CASSANDRA-15511
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15511
> Project: Cassandra
> Issue Type: Improvement
> Components: Local/Other
>Reporter: Benedict Elliott Smith
>Assignee: Benedict Elliott Smith
>Priority: Normal
>
> This patch utilises CASSANDRA-15510 to improve throughput and reduce garbage
> produced by a number of common operations, by employing
> {{transformAndFilter}}, {{transform}} and {{FastBuilder}}
> {{Row}}, {{Cell}} and {{ComplexColumnData}} cloning are implemented with
> {{BTree.transform}}, so no special builders are necessary; {{Rows.copy}}
> removed
> {{Rows.merge}} implemented using {{BTree.update}} and a {{ColumnData}}
> reconciler
> Zero-allocations if result of merge is same as {{existing}}
> Fewer comparisons
> {{ColumnData}} reconciler implemented in same manner
> {{Cells.reconcileComplex}} is retired
> {{ComplexColumnData}} reconciliation now
> Garbage-free if the merge has no effect
> Always fewer allocations
> Fewer comparisons
> {{FastBuilder}} employed widely:
> {{ClusteringIndexNamesFilter}} deserialization
> {{Columns}} deserialization
> {{PartitionUpdate}} deserialization
> {{AbstractBTreePartition}} construction
> Misc others
> The upshot of this work when combined with the proposed patch for
> CASSANDRA-15367 has a dramatic impact on operations over collection types -
> under contention, as much as 100x improved throughput, and hundreds of
> megabytes of reduced allocations. For all operations, allocations under
> contention and no contention are significantly reduced and throughput
> improved.
> I am still awaiting the final results of the comprehensive performance
> comparison for {{AtomicBTreePartition}}, after which I will establish what
> impact there might be on compaction and normal reads, both of which should be
> affected by this patch.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
