de.zib.scalaris.examples
Class FastStringBenchmark

java.lang.Object
  de.zib.scalaris.Benchmark
      de.zib.scalaris.examples.FastStringBenchmark

public class FastStringBenchmark
extends Benchmark

Mini benchmark of the Transaction and TransactionSingleOp class using custom objects provided by ErlangValueFastString and ErlangValueBitString.

Run the benchmark with java -cp scalaris-examples.jar de.zib.scalaris.examples.FastStringBenchmark

Since:

2.0

Version:

3.0

Author:

Nico Kruber, kruber@zib.de

Constructor Summary

FastStringBenchmark()

Method Summary

static void	main(String[] args) Runs a mini benchmark of the Transaction and TransactionSingleOp class using custom objects provided by ErlangValueFastString and ErlangValueBitString.
static void	minibench(int operations, int threadsPerNode, Set<Integer> benchmarks) Runs the benchmark.

Methods inherited from class de.zib.scalaris.Benchmark

getRandom, getRandom

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FastStringBenchmark

public FastStringBenchmark()

Method Detail

main

public static void main(String[] args)

Runs a mini benchmark of the Transaction and TransactionSingleOp class using custom objects provided by ErlangValueFastString and ErlangValueBitString. Accepts the same parameters as the Main.main(String[]) method's benchmark parameter.

Parameters:

args - command line arguments

See Also:

Main.main(String[])

minibench

public static void minibench(int operations,
                             int threadsPerNode,
                             Set<Integer> benchmarks)

Runs the benchmark. Tests some strategies for writing key/value pairs to scalaris:

1. writing OtpErlangBinary objects (random data, size = Benchmark.BENCH_DATA_SIZE)
2. writing OtpErlangString objects (random data, size = Benchmark.BENCH_DATA_SIZE)
3. writing String objects (random data, size = Benchmark.BENCH_DATA_SIZE)
4. writing String objects by converting them to OtpErlangBinarys (random data, size = Benchmark.BENCH_DATA_SIZE)

each with the given number of consecutive operations and parallel threads per Scalaris node,

• first using a new Transaction for each test,
• then using a new Transaction but re-using a single Connection,
• and finally re-using a single Transaction object.

Parameters:

operations - the number of test runs to execute

threadsPerNode - number of threads to spawn for each existing Scalaris node

benchmarks - the benchmarks to run (1-9 or -1 for all benchmarks)