About the performance comparison between Google JSV8 and Microsoft Jsrt

First of all, I did not get the actual comparison results, looking for interested friends can help to complete this comparison.

Benchmarksgame, offers a variety of language performance comparisons, but all for Linux under test, it's hard to compare Google with Microsoft stuff.

As we all know, JSV8 has excellent performance, but the specific and IE10/11 JS engine performance is lower, and there is no authoritative comparison results. From a browser-side comparison, Google does not use memory, regardless of power consumption, to improve the performance of chrome. and IE in more aspects for the sake of the user-system compatibility, low-carbon environmental protection, easy to expand and reuse has an absolute advantage. I specifically analyzed the Chrome rendering section, the default font is a bitmap font, not smooth processing, which will have a certain performance improvement, but loss of expressiveness. IE will display the text with the best results. Google in trickery!

The recent desire to embed a wide range of JavaScript into the program improves extensibility and the desire to get a thorough understanding of JSRT performance.

Benchmarks provides JSV8 's regular expression test code, which is representative of its performance over all languages. Copy the test code, implement the JSRT test, but did not provide the complete test source data, only found a 100KB source data. The table in the test page is the n value of 50000,500000,5000000, which does not understand where the data came from. Not in the source code.

I'm going to test it with 100KB of data, and the results are 108812 nanoseconds or so.

Benchmarksgame's data

Th>code b

n	cpu secs	elapsed secs	memory kb	≈ cpu load
50,000	0.05	0.05		373	17% 0% 0% 100%
500,000	0.35	0.39		373	5% 0% 0% 100%
5,000,000	3.29	3.30	484,640	373	0% 1% 1% 100%

This is the output of test results for Benchmarksgame 5000000:
AGGGTAAA|TTTACCCT 356
[CGT]GGGTAAA|TTTACCC[ACG] 1250
A[act]ggtaaa|tttacc[agt]t 4252
AG[ACT]GTAAA|TTTAC[AGT]CT 2894
Agg[act]taaa|ttta[agt]cct 5435
Aggg[acg]aaa|ttt[cgt]ccct 1537
AGGGT[CGT]AA|TT[ACG]ACCCT 1431
Agggta[cgt]a|t[acg]taccct 1608
agggtaa[cgt]| [Acg]ttaccct 2178

50833411
50000000
66800214

Here is my test program output:

Agggtaaa|tttaccct 0
[CGT]GGGTAAA|TTTACCC[ACG] 3
A[act]ggtaaa|tttacc[agt]t 9
AG[ACT]GTAAA|TTTAC[AGT]CT 8
Agg[act]taaa|ttta[agt]cct 10
Aggg[acg]aaa|ttt[cgt]ccct 3
AGGGT[CGT]AA|TT[ACG]ACCCT 4
Agggta[cgt]a|t[acg]taccct 3
agggtaa[cgt]| [Acg]ttaccct 5

101745
100000
133640

Feel my 100K data is a bit like benchmarksgame 50000 that data, also feel not.

Also look for a master to strike, test, or give advice.