Class TestSyncTimeRangeTracker
java.lang.Object
org.apache.hadoop.hbase.regionserver.TestSimpleTimeRangeTracker
org.apache.hadoop.hbase.regionserver.TestSyncTimeRangeTracker
-
Nested Class Summary
Modifier and TypeClassDescription(package private) static class
(package private) static class
-
Field Summary
Modifier and TypeFieldDescriptionstatic final HBaseClassTestRule
private static final int
private static final int
-
Constructor Summary
-
Method Summary
Modifier and TypeMethodDescriptionprotected org.apache.hadoop.hbase.regionserver.TimeRangeTracker
protected org.apache.hadoop.hbase.regionserver.TimeRangeTracker
getTimeRangeTracker
(long min, long max) void
Run a bunch of threads against a single TimeRangeTracker and ensure we arrive at right range.void
Run a bunch of threads against a single TimeRangeTracker and ensure we arrive at right range.Methods inherited from class org.apache.hadoop.hbase.regionserver.TestSimpleTimeRangeTracker
testAlwaysDecrementingSetsMaximum, testExtreme, testLegacySerialization, testRangeConstruction, testSerialization, testSimpleInRange, testTimeRangeInitialized, testTimeRangeTrackerNullIsSameAsTimeRangeNull
-
Field Details
-
CLASS_RULE
-
NUM_KEYS
- See Also:
-
NUM_OF_THREADS
- See Also:
-
-
Constructor Details
-
TestSyncTimeRangeTracker
public TestSyncTimeRangeTracker()
-
-
Method Details
-
getTimeRangeTracker
- Overrides:
getTimeRangeTracker
in classTestSimpleTimeRangeTracker
-
getTimeRangeTracker
protected org.apache.hadoop.hbase.regionserver.TimeRangeTracker getTimeRangeTracker(long min, long max) - Overrides:
getTimeRangeTracker
in classTestSimpleTimeRangeTracker
-
testArriveAtRightAnswer
Run a bunch of threads against a single TimeRangeTracker and ensure we arrive at right range. Here we do ten threads each incrementing over 100k at an offset of the thread index; max is 10 * 10k and min is 0.- Throws:
InterruptedException
-
testConcurrentIncludeTimestampCorrectness
Run a bunch of threads against a single TimeRangeTracker and ensure we arrive at right range. The data chosen is going to ensure that there are lots collisions, i.e, some other threads may already update the value while one tries to update min/max value.
-