Skip to content

#3926, set sourceRecordCount when updating statistics #626

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
yuexingri wants to merge 2 commits into from
Closed

#3926, set sourceRecordCount when updating statistics #626

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2b9108b
KYLIN-3926 set sourceRecordCount when updating statistics
kyotoYaho2025 Apr 1, 2019
5c727c1
minor, fix typo
yuexingri May 5, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions engine-mr/src/main/java/org/apache/kylin/engine/mr/common/CubeStatsReader.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -161,6 +161,10 @@ public int getSamplingPercentage() {
return samplingPercentage;
}

public long getSourceRowCount() {
return sourceRowCount;
}

public Map<Long, Long> getCuboidRowEstimatesHLL() {
return getCuboidRowCountMapFromSampling(cuboidRowEstimatesHLL, samplingPercentage);
}
Expand Down
5 changes: 5 additions & 0 deletions engine-mr/src/main/java/org/apache/kylin/engine/mr/common/CubeStatsWriter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,11 @@ public static void writeCuboidStatistics(Configuration conf, Path outputPath, //
writeCuboidStatistics(conf, outputPath, cuboidHLLMap, samplingPercentage, 0, 0, 0);
}

public static void writeCuboidStatistics(Configuration conf, Path outputPath, //
Map<Long, HLLCounter> cuboidHLLMap, int samplingPercentage, long sourceRecordCount) throws IOException {
writeCuboidStatistics(conf, outputPath, cuboidHLLMap, samplingPercentage, 0, 0, sourceRecordCount);
}

public static void writeCuboidStatistics(Configuration conf, Path outputPath, //
Map<Long, HLLCounter> cuboidHLLMap, int samplingPercentage, int mapperNumber, double mapperOverlapRatio,
long sourceRecordCoun) throws IOException {
Expand Down
14 changes: 13 additions & 1 deletion engine-mr/src/main/java/org/apache/kylin/engine/mr/steps/MergeStatisticsStep.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,6 +75,8 @@ protected ExecuteResult doWork(ExecutableContext context) throws ExecuteExceptio
try {

int averageSamplingPercentage = 0;
long sourceRecordCount = 0;
long effectiveTimeRange = 0;
for (String segmentId : CubingExecutableUtil.getMergingSegmentIds(this.getParams())) {
String fileKey = CubeSegment.getStatisticsResourcePath(CubingExecutableUtil.getCubeName(this.getParams()), segmentId);
InputStream is = rs.getResource(fileKey).content();
Expand All @@ -99,6 +101,13 @@ protected ExecuteResult doWork(ExecutableContext context) throws ExecuteExceptio
if (key.get() == 0L) {
// sampling percentage;
averageSamplingPercentage += Bytes.toInt(value.getBytes());
} else if (key.get() == -3) {
long perSourceRecordCount = Bytes.toLong(value.getBytes());
if (perSourceRecordCount > 0) {
sourceRecordCount += perSourceRecordCount;
CubeSegment iSegment = cube.getSegmentById(segmentId);
effectiveTimeRange += iSegment.getTSRange().duration();
}
} else if (key.get() > 0) {
HLLCounter hll = new HLLCounter(kylinConf.getCubeStatsHLLPrecision());
ByteArray byteArray = new ByteArray(value.getBytes());
Expand All @@ -120,8 +129,11 @@ protected ExecuteResult doWork(ExecutableContext context) throws ExecuteExceptio
tempFile.delete();
}
}
sourceRecordCount *= effectiveTimeRange == 0 ? 0 : newSegment.getTSRange().duration() / effectiveTimeRange;
averageSamplingPercentage = averageSamplingPercentage / CubingExecutableUtil.getMergingSegmentIds(this.getParams()).size();
CubeStatsWriter.writeCuboidStatistics(conf, new Path(CubingExecutableUtil.getMergedStatisticsPath(this.getParams())), cuboidHLLMap, averageSamplingPercentage);
CubeStatsWriter.writeCuboidStatistics(conf,
new Path(CubingExecutableUtil.getMergedStatisticsPath(this.getParams())), cuboidHLLMap,
averageSamplingPercentage, sourceRecordCount);
Path statisticsFilePath = new Path(CubingExecutableUtil.getMergedStatisticsPath(this.getParams()), BatchConstants.CFG_STATISTICS_CUBOID_ESTIMATION_FILENAME);
FileSystem fs = HadoopUtil.getFileSystem(statisticsFilePath, conf);
FSDataInputStream is = fs.open(statisticsFilePath);
Expand Down
2 changes: 1 addition & 1 deletion engine-mr/src/main/java/org/apache/kylin/engine/mr/steps/MergeStatisticsWithOldStep.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -120,7 +120,7 @@ protected ExecuteResult doWork(ExecutableContext context) throws ExecuteExceptio

String resultDir = CubingExecutableUtil.getMergedStatisticsPath(this.getParams());
CubeStatsWriter.writeCuboidStatistics(conf, new Path(resultDir), resultCuboidHLLMap,
averageSamplingPercentage);
averageSamplingPercentage, oldSegmentStatsReader.getSourceRowCount());

try (FSDataInputStream mergedStats = hdfs
.open(new Path(resultDir, BatchConstants.CFG_STATISTICS_CUBOID_ESTIMATION_FILENAME))) {
Expand Down