[SNAP-1034] Optimizations at Spark layer as seen in profiling #10
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 3 commits
September 7, 2016 12:14
- for all cases of implicit casts, convert to date or timestamp values instead of string when one side is a string - likewise when one side is a timestamp and other date then both are being converted to string; now convert date to timestamp
- added a aggBufferWithKeyAttributes to aggregates to be used to avoid nullable
checks in generated code in aggregate buffers used in HashAggregateExec (if aggregate
is on zero rows, then there will be no row in the map)
- use OpenHashMap in DictionaryEncoding which is faster than normal hash map;
added clear methods to OpenHashMap/OpenHashSet for reuse
- temp change to use a local cache for ClosureCleaner to avoid cleaning closures
that can be serialized as is
- minor correction in the string in HiveUtils
Reverting non-null aggregate attributes for Min, Max, First, Last since these depend on aggregate value to be null initially to set the value during iteration. Added a "initialValuesForGroup" like for aggregate attributes that will setup the value with zero for the data type for Sum, Average when creating initial aggregation buffer (instead of null) Renamed "aggregateBufferWithKeyAttribute" to "aggregateBufferAttributeForGroup" Reverted the optimization hack in ClosureCleaner for now.
ymahajan
pushed a commit
that referenced
this pull request
Jan 13, 2017
- added a aggBufferAttributeForGroup to aggregates to be used to avoid nullable
checks in generated code in aggregate buffers used in HashAggregateExec (if aggregate
is on zero rows, then there will be no row in the map); accompanying "initialValuesForGroup"
added for initial aggregation buffer values
- use OpenHashMap in DictionaryEncoding which is faster than normal hash map;
added clear methods to OpenHashMap/OpenHashSet for reuse
- minor correction in the string in HiveUtils
sumwale
pushed a commit
that referenced
this pull request
Jul 8, 2017
- added a aggBufferAttributeForGroup to aggregates to be used to avoid nullable
checks in generated code in aggregate buffers used in HashAggregateExec (if aggregate
is on zero rows, then there will be no row in the map); accompanying "initialValuesForGroup"
added for initial aggregation buffer values
- use OpenHashMap in DictionaryEncoding which is faster than normal hash map;
added clear methods to OpenHashMap/OpenHashSet for reuse
- minor correction in the string in HiveUtils
ymahajan
pushed a commit
that referenced
this pull request
Feb 22, 2018
- added a aggBufferAttributeForGroup to aggregates to be used to avoid nullable
checks in generated code in aggregate buffers used in HashAggregateExec (if aggregate
is on zero rows, then there will be no row in the map); accompanying "initialValuesForGroup"
added for initial aggregation buffer values
- use OpenHashMap in DictionaryEncoding which is faster than normal hash map;
added clear methods to OpenHashMap/OpenHashSet for reuse
- minor correction in the string in HiveUtils
Conflicts:
sql/core/src/main/scala/org/apache/spark/sql/execution/aggregate/HashAggregateExec.scala
sql/core/src/main/scala/org/apache/spark/sql/execution/joins/HashedRelation.scala
sql/core/src/main/scala/org/apache/spark/sql/execution/metric/SQLMetrics.scala
sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveUtils.scala
ashetkar
pushed a commit
that referenced
this pull request
Apr 5, 2018
* Use alpine and java 8 for docker images. * Remove installation of vim and redundant comment
sumwale
pushed a commit
to sumwale/spark
that referenced
this pull request
Nov 5, 2020
…oftware#10) - use OpenHashMap in DictionaryEncoding which is faster than normal hash map; added clear methods to OpenHashMap/OpenHashSet for reuse - add addLong/longValue methods to SQLMetric that use primitive longs instead of Long objects
sumwale
pushed a commit
that referenced
this pull request
Jul 11, 2021
- use OpenHashMap in DictionaryEncoding which is faster than normal hash map; added clear methods to OpenHashMap/OpenHashSet for reuse - add addLong/longValue methods to SQLMetric that use primitive longs instead of Long objects
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
Note that the closure cleaner change is just a temporary hack for testing. It will be turned into a proper shape by caching the steps for cleaning of each class, if any, and then applying those steps in order (this strategy may not work very well for polymorphic types but user can take care explicitly for such special cases). This is being tracked in a separate JIRA.
How was this patch tested?
Applied and tested with upstream spark branch-2.0