I think we need to add details like: which source we used. I'd also like a note one of the first pages, saying that this ESME version includes a Scala library that includes a known major bug involving memory allocation which detrimentally affected the test results. Just to avoid giving others the ability to take this report and say "Look. ESME is unstable and has poor performance."
We also need a simple description of the type of test that were performed - how many clients, etc. The link for the loadtester source code is also incorrect. Maybe for later tests we do capacity tests where you start with one user for minutes and then an additional user every 20 seconds up to 50 users and then have a ramp-down. We could combine this with more complicated test scenarios to get even better results. Otherwise, I think it is a great start and I can't wait to get a released version so that I can blog about it. D. -----Ursprüngliche Nachricht----- Von: Daniel Koller [mailto:[email protected]] Gesendet: Montag, 12. Jänner 2009 15:00 An: [email protected] Betreff: ESME Loadtest results for your review Hi all, I posted the first draft of the loadtest evaluation at http://esmeproject.googlecode.com/files/20090111_esmeloadtest_summary2.pdffor your review. Please give me feedback if you miss information in it, or if you think we should put clearer emphasis on some facts before communicate it "externally". Kind regards, Daniel -- --- Daniel Koller Jahnstrasse 20 80469 München * [email protected]
