On Wed, Jun 24, 2009 at 8:24 AM, Johan Svensson<[email protected]> wrote: > On Mon, Jun 22, 2009 at 4:37 PM, Macarse<[email protected]> wrote: >> On Mon, Jun 22, 2009 at 10:21 AM, Johan Svensson<[email protected]> >> wrote: >>> ... >>> You also run everything in the same transaction. I tried getting some >>> information on transaction and what isolation level the insert/query >>> you where running on the relational database was using but could not >>> see any info about it. If you split the create/insert part in one >>> transaction and run the query in a separate transaction the query >>> should drop down to about 350ms (that is with 128M heap and no >>> String.format). >> >> I thought that was a benefit for neo4j in the benchmark. >> Why separating inserts and queries in two transactions improves speed? >> > > Neo4j has been optimized for small updates/writes running together > with heavy concurrent reads. The reason it improves speed is that a > read only transaction does not have to apply any modifications that > has been done locally in the same transaction. > > Reading already committed (and cached) state is faster compared to > reading committed state together with uncommitted changes. > > Regards, > Johan > _______________________________________________ > Neo mailing list > [email protected] > https://lists.neo4j.org/mailman/listinfo/user >
Well, I finally gave the presentation on Thursday. It went really well. You can check it out here: http://code.google.com/p/grafos2009/downloads/list The presentation is in Spanish. Thanks for everyone's time and your quick replies! _______________________________________________ Neo mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
