Re: Load testing with JRobin and Newts

Hi Fred,

we observed those „holes“ in the graphs when by mistake the same system was defined in opennms more than once as a node.

It’s quite a difficult task you are trying – there are a lot of possibilities to optimize throughput using rrd. File caching by operating system and a lot of memory, disk block size, disabling read-ahead, number of write processes, enabling MNIO (rrd-configuration.properties), …

Just optimizing block size and read-ahead is a time consuming task – queued operations pending, completed updates, garbage collection, cpu usage, iowait, disk tps, disk io size, disk writes/reads, everything changes when you change the block size or read-ahead. So where or what is the optimum?

You should read those pages:
https://wiki.opennms.org/wiki/RRD_performance_fundamentals
https://oss.oetiker.ch/rrdtool-trac/wiki/TuningRRD

But it might be that the biggest advantage of using newts is to obtain good performance without the need to do a lot of performance tuning.

I would be very interested to hear about your results. I tried newts a few years ago, when it was still not integrated in OpenNMS, but hadn’t the time since to set up a real testbed, so the server is still running rrd (or, to tell the  truth,  jrb).


-Michael

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss