Le 30 sept. 2013 à 21:16, Les Mikesell <lesmikes...@gmail.com> a écrit :
> On Mon, Sep 30, 2013 at 1:22 PM, Fabrice Bacchella > <fbacche...@spamcop.net> wrote: >> >> Le 30 sept. 2013 à 19:37, Les Mikesell <lesmikes...@gmail.com> a écrit : >>> >>> Making the system use swap to emulate RAM for disk buffering is >>> clearly a bad idea. But, on a 2nd thought, I don't see how using >>> FILE can avoid any overhead added by linux filesystem access - NIO >>> must be (over)doing something to hold it in memory. >>> >> >> NIO is a little more than java.nio usage. It uses mmap internally throught >> MappedByteBuffer. But java hides to much of posix API, so it's not possible >> to use madvise to give hint to the kernel. That's where the difference with >> FILE is. >> > > > I just looked at http://code.google.com/p/rrd4j/ and saw that it can > use a MongoDB backend. Maybe the best answer for scalability is to > let someone else worry about it... > The currend db don't scale, it will be even worse without the pool. Each db is a record, is loaded once when the fill is open and written back at closure. The good scalable solution would be to use my new backend API with a cassandra back end. ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk _______________________________________________ Please read the OpenNMS Mailing List FAQ: http://www.opennms.org/index.php/Mailing_List_FAQ opennms-devel mailing list To *unsubscribe* or change your subscription options, see the bottom of this page: https://lists.sourceforge.net/lists/listinfo/opennms-devel
