Commit Graph

938 Commits

Author SHA1 Message Date
Xavier Léauté ad23e49777 use fixed-size mapdb cache to avoid heap growing uncontrollably 2014-12-05 15:34:50 -08:00
Xavier Léauté 7cd45a6e1f IncrementalIndex throws exception if limit exceeded
- For now uses a hardcoded ratio of aggregator to timeanddim buffer sizes
- canAppendRow is a workaround for realtime index since the
Firehose currently does not have a way of rolling back the last event in
case of error
- canAppendRow needs a fudge factor; there is a race between checking
if we can add a row and actually adding a row, because of the way MapDB
reports its size.
2014-12-04 14:38:16 -08:00
Xavier Léauté c7dbe6116c write byte data as is in smile 2014-12-04 10:57:56 -08:00
Xavier Léauté c21a82a697 upgrade LZ4 to operate directly on ByteBuffers 2014-12-04 10:57:56 -08:00
Xavier Léauté 0c521e0a77 update joda-time and fix min/max instant 2014-12-04 10:57:56 -08:00
nishantmonu51 269a51964e fix size calculation 2014-12-04 17:22:24 +05:30
nishantmonu51 4dc0fdba8a consider mapped size in limit calculation & review comments 2014-12-03 23:47:30 +05:30
nishantmonu51 da8bd7836b Introduce buffer size 2014-12-03 16:28:22 +05:30
fjy bc173d14fc a whole bunch of cleanup and fixes 2014-12-02 17:32:05 -08:00
nishantmonu51 b65933ffb8 make tests parameterised 2014-12-02 23:55:29 +05:30
nishantmonu51 6dc69c2f30 code cleanups & formatting 2014-12-02 22:44:33 +05:30
nishantmonu51 eac776f1a7 tests passing with on heap incremental index 2014-12-02 22:29:28 +05:30
Xavier Léauté 4eee7e69b9 fix cardinality aggregator caching 2014-11-26 15:00:37 -08:00
xvrl 5bc1be5ba0 Merge pull request #850 from metamx/druid-0.7.x-compressionstrategy
Compression strategy changes
2014-11-25 12:58:39 -08:00
Charles Allen c6043afa32 Removed empty function from CompressionStrategyTest 2014-11-25 12:57:06 -08:00
Charles Allen 6943db5251 Changed branching logic for LZFCompressor to return null only on error, and avoid checking in most circumstances 2014-11-25 12:53:11 -08:00
Charles Allen 9f945c2216 Removed lz4Fast from CompressedObjectStrategy for compression since it is not currently used 2014-11-24 16:11:03 -08:00
Charles Allen 70e3108282 Multiple speed improvements revolving around topN with HLL
Change serializer / deserializer for HyperLogLog
* Changed DirectDruidClient's InputStream handling. Is now ~10% faster for data heavy queries, and has lower variance in execution speed.
* Changed HLL Collector's toByteStream() method to be better optimized for small values. Is notably faster for small result quantities which fall into the sparse HLL bucket codepath.
  * No change for dense HLL which just uses a direct bytestream of the underlying byte data.

TopNNumericResultBuilder semi-aggressive loop unrolling for metricVals

Benchmark for HLL for sparse packing (small HLL bucket population):
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[0]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 216, GC.time: 0.42, time.total: 15.96, time.warmup: 0.22, time.bench: 15.74
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[1]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 217, GC.time: 0.45, time.total: 13.87, time.warmup: 0.02, time.bench: 13.85
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[2]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 55, GC.time: 0.16, time.total: 4.13, time.warmup: 0.00, time.bench: 4.12
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[3]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 55, GC.time: 0.16, time.total: 4.30, time.warmup: 0.00, time.bench: 4.30
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[4]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 8, GC.time: 0.03, time.total: 1.10, time.warmup: 0.00, time.bench: 1.09
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[5]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 8, GC.time: 0.03, time.total: 0.72, time.warmup: 0.00, time.bench: 0.72
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[6]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 1, GC.time: 0.00, time.total: 0.60, time.warmup: 0.00, time.bench: 0.60
HyperLogLogSerdeBenchmarkTest.benchmarkToByteBuffer[7]: [measured 100000 out of 100100 rounds, threads: 1 (sequential)]
 round: 0.00 [+- 0.00], round.block: 0.00 [+- 0.00], round.gc: 0.00 [+- 0.00], GC.calls: 2, GC.time: 0.01, time.total: 0.26, time.warmup: 0.00, time.bench: 0.25

Updates to HyperLogLogCollector toByteBuffer() based on code review

Removed changes from DirectDruidClient from this branch and put it in another branch.

Changed HyperLogLogCollector to have protected getters and setters

Remove unused ByteOrder from HyperLogLogCollector

Copyright header on HyperLogLogSerdeBenchmarkTest

Now with less ass!

Reformat in TopNNumericResultsBuilder. No code change

Removed unused import in HyperLogLogCollector

Replace AppendableByteArrayInputStream in DirectDruidClient
* Replace with SequenceInputStream fueled by an enumeration of ChannelBufferInputStream which directly wrap the response context ChannelBuffer

Modify TopNQueryQueryToolChest to use Arrays instead of Lists

Modify TopNQueryQueryToolChest to use Arrays instead of Lists

Revert accidental changes to DirectDruidClient

They should be in another merge request:
https://github.com/metamx/druid/pull/893

Fixes from code review
* Extracting names from AggregatorFactory classes now done with TopNQueryQueryToolChest.extractFactoryName
* Renamed variable in TopNNumericResultBuilder
2014-11-24 16:02:00 -08:00
fjy 13cae41f6c Merge branch 'master' into refactor-examples 2014-11-24 11:00:26 -08:00
xvrl 9ced097abd Merge pull request #895 from metamx/fix-interval-retry
A set of fixes to retry the query for missing segments in the timeline
2014-11-24 10:23:02 -08:00
fjy c88aff6205 remove unneeded constant 2014-11-24 10:20:02 -08:00
fjy 9da66291e1 change naming to common config 2014-11-21 15:36:42 -08:00
Charles Allen 8f80d9e189 Update CompressedObjectStrategy to try-with-resources but print log error if error while closing 2014-11-21 11:35:11 -08:00
Charles Allen fc9a54ea48 Fix CompressedObjectStrategy LZFCompressor to ignore error on close of ResourceHolder 2014-11-21 10:49:43 -08:00
Charles Allen f8ce68565b Modified CompressedObjectStrategy to use 0xFF for Uncompressed 2014-11-21 10:33:53 -08:00
Charles Allen aa49e56ed6 Merge remote-tracking branch 'origin/master' into druid-0.7.x-compressionstrategy 2014-11-21 10:29:40 -08:00
fjy ef62bccdec ignore benchmark 2014-11-20 16:52:19 -08:00
nishantmonu51 e3260aa177 Filtered Aggregator fixes + enhancements
- fix NPE on IncrementIndex
- refactor code to support AND, OR filter
- tests for AND & OR filter
- handling for missing column / null values
2014-11-20 15:17:18 -08:00
fjy 47f5c1bd0a fix retry interval is stupid 2014-11-20 12:50:56 -08:00
fjy 3d9d989a9f A set of fixes to retry the query for missing intervals in the timeline 2014-11-20 12:04:37 -08:00
nishantmonu51 0ab34f86da Revert "fix filtered Aggregator"
This reverts commit 6fd37ce023.
2014-11-20 10:17:01 +05:30
nishantmonu51 6fd37ce023 fix filtered Aggregator
fix filtered Aggregator
remove unused name parameter for filtered aggregator
add tests
2014-11-20 09:29:26 +05:30
fjy a49e673122 put back another missing test 2014-11-19 16:55:20 -08:00
fjy 14668846aa add back some tests 2014-11-19 14:35:26 -08:00
fjy fdeab0c6af make Druid case sensitive 2014-11-19 14:27:31 -08:00
Fangjin Yang 590d31799e Merge pull request #876 from metamx/remove-backwards-compatible
Remove backwards compatible
2014-11-19 14:33:14 -07:00
Charles Allen 18f44beee9 CompressedObjectStrategy improvements
* Added more unit tests
* Now properly uses safe / fast decompressor for LZ4
* Now chooses fastest lz4 instance instead of only looking at Java implmentations
* Encapsulate ResourceHolder in try-with-resources to make sure they close correctly
2014-11-19 11:10:59 -08:00
Charles Allen ccc757dc64 Merge remote-tracking branch 'origin/master' into druid-0.7.x-compressionstrategy 2014-11-19 09:39:35 -08:00
Charles Allen 1bbc8fcbe5 Allow Smile to fall back to text
* Modify SmileFactory to set the delegate to text option.
  * This option only occurs when a Reader type object is passed in to the deserialization stuff
  * This is needed by the X-Druid-Response-Context header return value, which is JSON
2014-11-18 15:16:14 -08:00
Charles Allen 42517f5d37 Merge pull request #884 from metamx/optimize-topN-pruning
optimise pruning of aggs
2014-11-18 14:19:30 -08:00
xvrl a96eaeb036 Merge pull request #882 from metamx/now_with_OPEN_SOURCE
Added src jar build to maven poms and re-formatted to conform to style guidelines.
2014-11-18 13:00:04 -08:00
nishantmonu51 6023d602e6 optimise pruning of aggs
optimise pruning of aggregators for topN
2014-11-19 00:17:25 +05:30
Charles Allen dc66e1708e Added src jar build to maven poms and re-formatted to conform to style guidelines. 2014-11-18 09:05:30 -08:00
Xavier Léauté d914afe1cd make defaultVersion configurable for non-jar testing 2014-11-17 13:54:32 -08:00
nishantmonu51 0c2d06475d merge from master 2014-11-17 19:19:18 +05:30
fjy d5c4282766 fix broken ut 2014-11-14 13:45:42 -08:00
xvrl e1e171ad20 Merge pull request #865 from metamx/fix-retry-qr
Fix a couple of bugs with retry query runner after testing it locally
2014-11-14 13:33:51 -08:00
fjy df1ad95075 remove useless config 2014-11-14 13:32:19 -08:00
fjy d641d41f9e address another cr 2014-11-14 13:29:59 -08:00
fjy 7736c3fc27 address cr 2014-11-14 13:28:32 -08:00