public static class BenchmarkJournalWriteRate.BenchmarkSustainedTransferOptimium extends BenchmarkJournalWriteRate.AbstractBenchmarkOptimium
Writes the same amount of data using a single nio "write buffer" operation on a pre-extended file. The buffer is a direct buffer, so it is allocated in the OS memory. The write should be pure sequential IO. This case should produce the "best-case" optimium throughput to disk for sustained IO. The journal SHOULD NOT be able approach this best case scenario. Comparison to this case should reveal the overhead of the journal, Java, and block-oriented IO when compare to sustained sequential data transfer from RAM to disk. Since block-based IO is, in fact, better, one can only presume that the nio library has some problem with very large writes.
Note: This overrides several methods in the base class in order to conduct a test without the use of a journal.
Note: I have seen block-based IO perform better in cases where system resources were low (the disk was nearly full).
TestCase2.MyProperties, TestCase2.RandomType_randomType, log| Constructor and Description |
|---|
BenchmarkJournalWriteRate.BenchmarkSustainedTransferOptimium() |
| Modifier and Type | Method and Description |
|---|---|
int |
getRecordSize()
The entire extent in one sustained write.
|
doOptimiumWriteRateTest, getFilename, getInitialExtent, setUp, tearDown, testNoIsolationassertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEqualsWithinUlps, assertSameArray, assertSameArray, assertSameBigDecimal, assertSameBigDecimal, assertSameBigInteger, assertSameBigInteger, assertSameIterator, assertSameIterator, assertSameIteratorAnyOrder, assertSameIteratorAnyOrder, assertSameValue, assertSameValue, assertZeroUlps, assertZeroUlps, fail, getInnerCause, getNormalInt, getProjectBuildPath, getProperties, getRandomObject, getRandomObject, getRandomOrder, getRandomString, getTestInputStream, getTestResource, getTestResource, getUlps, getUlps, isDEBUG, isDEBUG, isINFO, isINFO, isInnerCause, logPropertiesassertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertFalse, assertFalse, assertNotNull, assertNotNull, assertNotSame, assertNotSame, assertNull, assertNull, assertSame, assertSame, assertTrue, assertTrue, countTestCases, createResult, fail, fail, failNotEquals, failNotSame, failSame, format, getName, run, run, runBare, runTest, setName, toStringpublic BenchmarkJournalWriteRate.BenchmarkSustainedTransferOptimium()
public int getRecordSize()
getRecordSize in class BenchmarkJournalWriteRate.AbstractBenchmarkOptimiumCopyright © 2006–2019 SYSTAP, LLC DBA Blazegraph. All rights reserved.