Hi John, I used simple apache benchmark with a random selected css. ab -n 2000 -H "Accept-Encoding: gzip, deflate" http://172.17.10.10/client/css/cloudstack3.css
This looks big enough, it is about 200 K uncompressed. I used the configuration you published in your blog and I disabled the static resource compression by commenting the servlet out from the web.xml, I verified the configurations by examining the headers and message body. Since the tested VM is a local VM, the network latency is really close to zero. Please share your results! I was surprised first about the low performance of dynamic compression, I expected it to be somewhat slower and I believe the resource size or maybe even the content itself can alter the results. But theoretically it can never be as fast as static compression since that is just plain IO. Some other comparisons done by others http://www.webperformance.com/library/reports/moddeflate/ http://www.webperformance.com/library/reports/iis7_compression/ - ISS too As I wrote, I believe the best solution would be combining both static and dynamic. On Fri, Mar 21, 2014 at 10:03 PM, John Kinsella <j...@stratosec.co> wrote: > Canya tell us a little more about the test you're doing? What URL are you > fetching, how many times etc. Just curious to tinker myself this weekend if > I have some time. :) > > On Mar 21, 2014, at 1:07 PM, Laszlo Hornyak <laszlo.horn...@gmail.com> > wrote: > > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/12228/#review38181 > ----------------------------------------------------------- > > > > https://docs.google.com/spreadsheet/ccc?key=0ApMkIX1Ygx8zdHhqV2RETy05SDU1WER3Z2JEN3ktZHc&usp=sharing > > Comparison of 3 configurations, the dynamic compression solution is the > last one. This test was conducted in a VM with 2 vCPU and 4 GB RAM on a > AMD E2-1800. > > > - Laszlo Hornyak > > > On March 20, 2014, 7:53 p.m., Laszlo Hornyak wrote: > > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/12228/ > ----------------------------------------------------------- > > (Updated March 20, 2014, 7:53 p.m.) > > > Review request for cloudstack, Brian Federle, Darren Shepherd, and > Prasanna Santhanam. > > > Repository: cloudstack-git > > > Description > ------- > > CloudStack at first use downloads some 3.5 MB of css and javascript to the > client. With a weak internet connection, this might take a long time. With > gzip compression content can be compressed to 850 KB. > > This version of the patch uses a custom plugin to compress static > resources, so that no dynamic compression is needed at runtime. When the > static resource servlet notices that there is gzipped version of the > resource and the client accepts gzipped content, then it is going to send > the gziped version, while still respects http caching. > > > Diffs > ----- > > client/WEB-INF/web.xml 1af38e1 > client/pom.xml d8dbde7 > server/src/com/cloud/servlet/StaticResourceServlet.java PRE-CREATION > server/test/com/cloud/servlet/StaticResourceServletTest.java PRE-CREATION > > Diff: https://reviews.apache.org/r/12228/diff/ > > > Testing > ------- > > yes, tested with firefox and chrome > > > Thanks, > > Laszlo Hornyak > > > > > Stratosec <http://stratosec.co/> - Compliance as a Service > o: 415.315.9385 > @johnlkinsella <http://twitter.com/johnlkinsella> > > -- EOF
