Hi, can you perhaps elaborate a bit on the type of datasets? You mention it's a tree, but how large is this tree?
Kind regards, Jeroen On Wed, Sep 25, 2013 at 1:48 PM, Seidel. Robert <[email protected]>wrote: > Hi, > > I'm having a little trouble with the write performance of Jackrabbit. I > use a bundled database persistence manager with either Oracle or MS SQL > Server as database. My application is currently limited to three datasets > per second, which are stored in a tree structure within jackrabbit - for my > superiors this is not enough, they need ten datasets per second or more. > > The profiler shows me, that most of the time is consumed by database > operations, inserts in jackrabbit tables, selects and commit. Is there any > possibility to improve the performance here? Multithreading doesn't help, > cause the journal implementation blocks the other threads. More processes > doesn't help either. > > How do you handle situations were much data should be inserted in a > jackrabbit repository? > > Kind regards, > > Robert > ________________________________ > > Treffen Sie AEB auf dem DLK > Netzwerken Sie mit AEB im Oktober auf dem 30. Deutschen Logistik Kongress > | 23.-25. Oktober 2013 | Berlin > Weitere Informationen unter www.aeb.de/dlk< > http://www.aeb.de/de/events/dlk.php>. > -- Jeroen Reijn Solution Architect Hippo Amsterdam - Oosteinde 11, 1017 WT Amsterdam Boston - 101 Main Street, Cambridge, MA 02142 US +1 877 414 4776 (toll free) Europe +31(0)20 522 4466 www.onehippo.com http://about.me/jeroenreijn
