MapBench results published (ductus, pisces vs marlin 0.3)

[bourges.laurent]

Here are the latest benchmark results:
http://htmlpreview.github.io/?https://github.com/bourgesl/mapbench/blob/master/results/report/compare/compareResults.htm

Marlin 0.3 is the winner:
- 60% faster than pisces for small workload; 300-400% faster for high workload
- 1 thread: 1% slower than ductus for small workload; 90% faster for high workload
- 4 thread: 90% faster than ductus for small workload; 400% faster for high workload

MapBench runs made with the following setup:
- Oracle JDK8 b121 for ductus
- OpenJDK8 b65 for pisces / marlin 0.3
- linux 64 OS (fedora 14) on my dual-core i7 (+HT) laptop
- JVM settings = "-server -Xms2048m  -Xmx2048m"
- warmup phase using 4 threads and 200 iterations

Laurent

[Andrea Aime]

Hi Laurent,

thanks a lot for the summary, I'm writing a blog post to show the combined results of Marlin

and GeoServer under a couple of WMS workload, the above makes for an interesting introduction,

first show how much faster the rasterizer is, and then show how that translates in a couple

of full WMS benchmarks.

Once I'm done writing the blog I'll send you a private copy, in

case you'd like to review it before the publication (which is hopefully going to happen Monday morning)

Cheers

Andrea

--

== Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information ==

Ing. Andrea Aime 

@geowolf

Technical Lead

GeoSolutions S.A.S.

Via Poggio alle Viti 1187

55054  Massarosa (LU)

Italy

phone: +39 0584 962313

fax: +39 0584 1660272

mob: +39  339 8844549

http://www.geo-solutions.it

http://twitter.com/geosolutions_it

-------------------------------------------------------