Distributed Query Optimizations: FastAggregate and FastCombine #59
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
…ased on the number ParameterSets that it gets in dispatchWorkFragment(). This causes problems in VoltSystemProcedure.executeSysProcPlanFragmentsAsync() because it is making multiple ParameterSets per fragment... I need to sort this out in order to get @SnapshotSave working...
…hich Statement invocation in LocalTransaction. Still working through some problems to get this work. Not sure if this solves our problem...
…or preventing ArgumentParser from updating the catalog
…'t seem to work in the PartitionExecutor though...
…entId arrays. We're good to go with this optimization
…ns. QueryCache.addTransactionQueryResult() is now thread-safe. Minor tweaks for debugging output
…tion needed by a txn. This was broken out into a separate object to minimize the number of new fields that we need in AbstractTransaction for this feature. HStoreObjectPools will not create the MapReduce or PrefetchState pools if the catalog doesn't have any procedures that use those features
…ameter to use VoltDB's @ProcInfo catalog information to determine whether txns are single-partition or not.
…PartitionEstimator if the embedded basePartition not set. We now make sure that we set our in-memory HStoreConf when we call VoltServerConfig.setConfParameter()
…r to retrieve prefetch query results and prevent us from having to send a network message!! Of course this only works if the partitions are in the same HStoreSite but it's a start...
…) instead of in its constructor. This should help lower the amount of memory needed in the regression suite tests
… of the invocations throttled.
…g with Charles...(I'm not blaming him) Now I'm going to home have pasta for the third day in a row...
…voke the JVM's garbage collector. Not sure if this is going to make a difference...
… seeking to the row that we're trying to add when we can reference it directly using the active row
…-line tool. Added support for @adhoc
…invoke'. It's a hack, but oh well
…PlanFragment ids with certain parameters encoded in them
Compression added to WAL, merging back to diagnose ClosedChannelException
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in 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.
Add fastaggregate_sum and fastcombine feature in Java.
Add fragmentIds[] loop and use fragmentId to check fastaggregate or fastcombine planfragments in PartitionExecutor, delete some unnecessary object creation, add check before log debug, add groupby check in FastAggregateOptimization, change executeFastAggregate() and executeFastCombine() to execute(). And add @author comments in the classes which I change or create