Hello All, I'm writing to share the initial results of a DSpace stress test we are performing, and to ask for your thoughts and suggestions as we begin ramping up for the next round of tests.
http://sites.google.com/a/ohiolink.edu/drmc/Home/stretch-armstrong Background: At OhioLINK we've built a federation of DSpace instances across the state called the Digital Resource Commons (http://drc.ohiolink.edu), and we hope to expand our offerings beyond the academic library community. As part of that research, we are building a multi-million record test of DSpace using Amazon's Elastic Compute Cloud (EC2). Item Import: Like many of you, we noticed longer batch submission times as our repository grew beyond 200,000 items and knew we needed to find a solution. Our first goal was to confirm the results of the ROAD Project (http://www.jisc.ac.uk/whatwedo/programmes/reppres/tools/road.aspx) test mentioned by Stewart Lewis. In Stewart's scenario, the entire 300,000 record submission took place at one time. We wanted to see if the problem exists even when the submission is broken up into several smaller blocks and takes place over a period of days. Our initial data confirm and extend the results Stewart posted, 'DSpace at a Third of a Million Items.' http://blog.stuartlewis.com/wp-content/uploads/2009/01/dspace-banding.png While he was interested in the performance of the SWORD client, his experiment shows a steady increase in the time-to-ingest for a single batch submission of 300,000 records. We've confirmed the problem rests with the submission process itself, and is not just an issue with extremely large 'one shot' batches. Special thanks to Stewart (Auckland University), Tom De Mulder and Simon Brown (Cambridge University) for their early comments and assistance. Please feel free to send along your own insights and suggestions, John Davison Assistant Director Digital Resource Development ------------------------------------------------------------------------------ Crystal Reports - New Free Runtime and 30 Day Trial Check out the new simplified licensing option that enables unlimited royalty-free distribution of the report engine for externally facing server and web deployment. http://p.sf.net/sfu/businessobjects _______________________________________________ DSpace-tech mailing list DSpace-tech@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspace-tech
