Are the Java 6 -server programs faster?
Do the Java 6 -server programs use optimized assembly code libraries? Are they small simple programs or very optimized programs? Do the Java 6 -server programs use a fraction of the time used by others, or do they use several times more?
| Java 6 -server used what fraction? used how many times more? | ||||
|---|---|---|---|---|
| Programs | Time | Memory | Code | Reduced N |
| ↓ n-body | 1/4 | 18× | ± | |
| ↓ sum-file | 1/2 | 14× | 2× | |
| ↓ mandelbrot | 1/2 | 15× | ± | |
| ↓ spectral-norm | ± | 8× | ± | |
| ↓ recursive | ± | 11× | ± | |
| ↓ nsieve | ± | 2× | ± | |
| ↓ nsieve-bits | ± | 3× | 3× | |
| ↓ k-nucleotide | ± | 2× | ± | |
| ↓ reverse-complement | ± | ± | ± | |
| ↓ fannkuch | ± | 15× | ± | |
| ↓ regex-dna | ± | 2× | ± | |
| ↓ binary-trees | ± | 3× | ± | |
| ↓ partial-sums | 2× | 15× | ± | |
± look at the measurements and then look at the programs.
Are the Java 6 -server programs faster?
| Program & Logs | CPU secs | Elapsed secs | Memory KB | Code B | ~ CPU Load |
|---|---|---|---|---|---|
| n-body | |||||
| Java 6 -server | 14.75 | 11,524 | 1424 | ||
| OCaml | 54.00 | 644 | 1239 | ||
| sum-file | |||||
| Java 6 -server | 4.10 | 9,972 | 226 | ||
| OCaml | 8.74 | 692 | 138 | ||
| mandelbrot | |||||
| Java 6 -server | 3.24 | 11,092 | 623 | ||
| OCaml | 6.06 | 716 | 444 | ||
| spectral-norm | |||||
| Java 6 -server | 24.01 | 10,008 | 514 | ||
| OCaml | 31.63 | 1,196 | 369 | ||
| recursive | |||||
| Java 6 -server | 6.76 | 12,044 | 427 | ||
| OCaml | 7.47 | 1,120 | 389 | ||
| nsieve | |||||
| Java 6 -server | 2.25 | 20,592 | 296 | ||
| OCaml | 2.16 | 9,380 | 295 | ||
| nsieve-bits | |||||
| Java 6 -server | 5.04 | 13,364 | 523 | ||
| OCaml | 4.41 | 5,020 | 176 | ||
| k-nucleotide | |||||
| Java 6 -server | 15.89 | 70,964 | 1052 | ||
| OCaml | 13.83 | 34,636 | 845 | ||
| reverse-complement | |||||
| Java 6 -server | 1.30 | 58,932 | 592 | ||
| OCaml | 1.02 | 47,244 | 540 | ||
| fannkuch | |||||
| Java 6 -server | 11.05 | 8,884 | 555 | ||
| OCaml | 8.30 | 592 | 485 | ||
| regex-dna | |||||
| Java 6 -server | 7.81 | 75,892 | 921 | ||
| OCaml | 5.54 | 45,364 | 615 | ||
| binary-trees | |||||
| Java 6 -server | 6.89 | 26,808 | 603 | ||
| OCaml | 4.83 | 8,732 | 469 | ||
| partial-sums | |||||
| Java 6 -server | 9.15 | 8,488 | 474 | ||
| OCaml | 4.64 | 564 | 508 | ||
| chameneos-redux | |||||
| Java 6 -server | 17.51 | 12,092 | 1429 | ||
| fasta | |||||
| Java 6 -server | 21.46 | 9,184 | 1240 | ||
| pidigits | |||||
| Java 6 -server | 1.92 | 8,992 | 938 | ||
Java 6 -server : ubiquitous jit server virtual machine
java version "1.6.0_07"
Java(TM) SE Runtime Environment (build 1.6.0_07-b06)
Java HotSpot(TM) Server VM (build 10.0-b23, mixed mode, sharing)
Home Page: http://java.sun.com/j2se/
Download: http://java.sun.com/j2se/downloads/
"Remember how HotSpot works. It starts by running your program with an interpreter. When it discovers that some method is "hot" -- that is, executed a lot, either because it is called a lot or because it contains loops that loop a lot -- it sends that method off to be compiled. After that one of two things will happen, either the next time the method is called the compiled version will be invoked (instead of the interpreted version) or the currently long running loop will be replaced, while still running, with the compiled method. The latter is known as "on stack replacement" and exists in the 1.3/1.4 HotSpot based systems."
Benchmarking the Java HotSpot VM
