[kepler-dev] [Bug 5403] Improve the performance of the workflow archiving data from dataturbine to metacat

bugzilla-daemon at ecoinformatics.org bugzilla-daemon at ecoinformatics.org
Fri Aug 24 17:49:54 PDT 2012


http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5403

--- Comment #9 from Derik Barseghian <barseghian at nceas.ucsb.edu> 2012-08-24 17:49:54 PDT ---
Today I did a bunch of testing of the new (archiveModified.xml -- attached) and
the existing (archiveDataturbineDataToMetacat.kar) workflows, with and without
the EcogridWriter changes, trying to track down why run times were dismally
long.

The summary is uploading to dev2 seems to take a minimum of ~125s per
datapackage right now, no matter the workflow, or EcogridWriter actor changes.
Uploading to chico1, which is further away and a slower machine, for similar
amounts of data, takes only ~30s to 40s per datapackage. So we'll have to look
into why dev2 is performing so badly later. Jing restarted the tomcat, it
didn't help.

So i'm planning to include the EcogridWriter changes and the new version of the
workflow in the sensor-view release. I've spliced in the nice end display
results that Jing created for the old workflow.

-- 
Configure bugmail: http://bugzilla.ecoinformatics.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA Contact for the bug.


More information about the Kepler-dev mailing list