Java 7 vs Scala | Computer Language Benchmarks Game

1 : Are the Java 7 programs faster? At a glance.

Each chart bar shows, for one unidentified benchmark, how much the fastest Java 7 program used compared to the fastest Scala program.

These are not the only compilers and interpreters. These are not the only programs that could be written. These are not the only tasks that could be solved. These are just 10 tiny examples.

2 : Are the Java 7 programs faster? Approximately.

Each table row shows, for one named benchmark, how much the fastest Java 7 program used compared to the fastest Scala program.

Java 7 used what fraction? used how many times more?
Benchmark	Time	Memory	Code
fasta-redux	¹/₂	¹/₂	±
regex-dna	¹/₂	±	±
binary-trees	±	±	2×
fasta	±	±	±
reverse-complement	±	±	2×
fannkuch-redux	±	±	±
mandelbrot	±	¹/₃	±
n-body	±	±	±
spectral-norm	±	±	±
pidigits †	±	±	2×
k-nucleotide	2×	2×	±

Java 7 used what fraction? used how many times more?
Time-used	\|-	\|---	25%	median	75%	---\|	-\|
(Elapsed secs)	¹/₂	¹/₂	±	±	±	±	2×

† possible mismatch - one-core program compared to multi-core program.

± read the measurements and then read the program source code.

3 : Are the Java 7 programs faster? Measurements.

These are not the only tasks that could be solved. These are just 10 tiny examples. These are not the only compilers and interpreters. These are not the only programs that could be written.

For each named benchmark, measurements of the fastest Java 7 program are shown for comparison against measurements of the fastest Scala program.

Program Source Code	CPU secs	Elapsed secs	Memory KB	Code B	≈ CPU Load
fasta-redux
Java 7	2.56	2.52	17,308	1443	0% 1% 1% 100%
Scala	5.00	4.70	33,916	1372	5% 1% 100% 2%
regex-dna
Java 7	24.82	8.55	709,672	929	65% 82% 65% 79%
Scala	34.01	15.48	980,808	668	39% 77% 61% 43%
binary-trees
Java 7	22.31	9.24	523,784	1007	37% 65% 58% 83%
Scala	16.37	12.46	477,616	494	12% 12% 11% 99%
fasta
Java 7	5.21	5.14	18,284	1507	0% 1% 1% 100%
Scala	6.40	6.29	24,720	1053	0% 1% 1% 100%
reverse-complement
Java 7	2.79	1.31	298,404	1661	61% 77% 40% 39%
Scala	1.97	1.52	439,420	761	4% 3% 92% 35%
fannkuch-redux
Java 7	51.90	13.15	17,348	1282	99% 99% 98% 98%
Scala	55.35	14.13	25,604	1017	98% 97% 99% 98%
mandelbrot
Java 7	27.82	7.07	69,460	796	99% 98% 99% 98%
Scala	28.57	7.43	233,828	786	96% 96% 96% 98%
n-body
Java 7	22.51	22.50	17,256	1424	0% 0% 100% 0%
Scala	23.52	23.47	23,708	1358	1% 2% 1% 100%
spectral-norm
Java 7	17.61	4.51	17,540	950	97% 97% 99% 98%
Scala	17.75	4.63	23,308	720	95% 95% 95% 99%
pidigits
Java 7	10.20	4.29	60,520	1808	68% 62% 65% 45% †
Scala	3.56	2.98	52,776	1125	23% 11% 79% 7% †
k-nucleotide
Java 7	46.39	13.25	450,972	1630	86% 84% 96% 85%
Scala	23.86	8.31	215,304	2080	56% 79% 75% 77%

† possible mismatch - one-core program compared to multi-core program.

4 : Are there other Java 7 programs for these benchmarks?

Remember - those are just the fastest Java 7 and Scala programs measured on this OS/machine. Check if there are other implementations of these benchmark programs for Java 7.

Maybe one of those other Java 7 programs is fastest on a different OS/machine.

5 : Are there other faster programs for these benchmarks?

Remember - those are just the fastest Java 7 and Scala programs measured on this OS/machine. Check if there are faster implementations of these benchmark programs for other programming languages.

Maybe one of those other programs is fastest on a different OS/machine.

Java 7 : ubiquitous jit server virtual machine

java version "1.7.0_11"
Java(TM) SE Runtime Environment (build 1.7.0_11-b21)
Java HotSpot(TM) 64-Bit Server VM (build 23.6-b04, mixed mode)

Home Page: Java SE at a Glance

Download: Java SE Downloads

Let's see how much, or how little, the time taken to invoke the JVM might contribute to the usual Java program times shown in the benchmarks game. Here are some additional (Intel® Q6600® quad-core) elapsed time measurements, taken after the Java programs started and before they exited.

In the first case (Cold), we simply started and measured the program 66 times; and then discarded the first measurement leaving 65 data points.

   public static void main(String[] args){
      for (int i=0; i<1; ++i){ 
         System.gc(); 
         long t1 = System.nanoTime();
         nbody.program_main(args);
         long t2 = System.nanoTime();
         System.err.println( String.format( "%.6f", (t2 - t1) * 1e-9 ) );         
      }
   }

In the second case (Warmed), we started the program once and repeated measurements again and again and again 66 times without restarting the JVM; and then discarded the first measurement leaving 65 data points.

   public static void main(String[] args){
      for (int i=0; i<66; ++i){ 
         System.gc(); 
         long t1 = System.nanoTime();
         nbody.program_main(args);
         long t2 = System.nanoTime();
         System.err.println( String.format( "%.6f", (t2 - t1) * 1e-9 ) );         
      }
   }

Compare these additional measurements against the usual Java program measurements shown in the benchmarks game --

"1.7.0_06" Java HotSpot(TM) 64-Bit Server VM
System.nanoTime()	1) Cold		2) Warmed
	mean	σ	mean	σ	usual
meteor contest	0.0118s	0.0007	0.0016s	0.0002	0.22s
fasta-redux	2.45s	0.00	2.32s	0.00	2.51s
spectral-norm	4.44s	0.02	4.20s	0.16	4.51s
pidigits	4.69s	0.09	4.44s	0.05	4.61s
fasta	5.07s	0.46	4.84s	0.02	5.13s
chameneos-redux	5.84s	0.46	5.70s	0.48	5.65s
mandelbrot	7.93s	0.23	7.99s	0.01	7.02s
k-nucleotide	8.09s	0.28	--	--	8.05s
regex-dna	8.65s	0.27	--	--	8.61s
binary-trees	10.54s	0.28	7.66s	0.16	9.08s
fannkuch-redux	16.89s	1.32	17.26s	0.10	17.38s
nbody	22.43s	0.00	22.41s	0.00	22.50s
binary-trees-redux	34.15s	0.39	33.93s	0.31	33.38s

The largest and most obvious effects of bytecode loading and dynamic optimization can be seen with the meteor-contest program which only runs for a fraction of a second.

Home Conclusions License Play