natas_hw
/
cassandra
mirror of https://github.com/apache/cassandra
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Extract calculateSplitCount method 

Also added characterization tests

Co-Authored-By: Danny Faught <13002425+swalchemist@users.noreply.github.com>
This commit is contained in:
Lada Kesseler 2024-09-20 13:20:45 -07:00
parent fe025c7f79
commit 2d950d5bd0
2 changed files with 34 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
src/java/org/apache/cassandra/service/StorageService.java
View File

@ -3546,15 +3546,24 @@ public class StorageService extends NotificationBroadcasterSupport implements IE
long totalRowCountEstimate = cfs.estimatedKeysForRange(range);
// splitCount should be much smaller than number of key samples, to avoid huge sampling error
int minSamplesPerSplit = 4;
int maxSplitCount = keys.size() / minSamplesPerSplit + 1;
int splitCount = Math.max(1, Math.min(maxSplitCount, (int)(totalRowCountEstimate / keysPerSplit)));
int splitCount = calculateSplitCount(keysPerSplit, totalRowCountEstimate, keys.size());
List<Token> tokens = keysToTokens(range, keys);
return getSplits(tokens, splitCount, cfs);
}
static int calculateSplitCount(int keysPerSplit, long totalRowCountEstimate, int numberOfKeys)
{
// splitCount should be much smaller than number of key samples, to avoid huge sampling error
int minSamplesPerSplit = 4;
int maxSplitCount = numberOfKeys / minSamplesPerSplit + 1;
int splitCount = Math.max(1,
Math.min(
maxSplitCount,
(int) (totalRowCountEstimate / keysPerSplit)));
return splitCount;
}
private List<Pair<Range<Token>, Long>> getSplits(List<Token> tokens, int splitCount, ColumnFamilyStore cfs)
{
double step = (double) (tokens.size() - 1) / splitCount;

21
test/unit/org/apache/cassandra/service/StorageServiceTest.java
View File

@ -316,4 +316,25 @@ public class StorageServiceTest extends TestBaseImpl
assertEquals("Cannot specify tokens without keyspace.", ex.getMessage());
}
}
@Test
public void calculateSplitCountHappyPath()
{
int result = StorageService.calculateSplitCount(2, 4, 40);
assertEquals(2, result);
}
@Test
public void calculateSplitCountForMin()
{
int result = StorageService.calculateSplitCount(2, 100, 40);
assertEquals(11, result);
}
@Test
public void calculateSplitCountForMaxWithOverflow()
{
int result = StorageService.calculateSplitCount(1, Long.MAX_VALUE, 4);
assertEquals(1, result);
}
}