What to Do with Data Captured from Monitoring
A while back I talked about monitoring you should perform on your systems. It doesn’t matter if those systems are database servers, middleware, or something different entirely. The point is that you need to monitor those servers so that you know what typical operation parameters are. With that baseline, when performance degrades, you are more likely to locate the cause.
Today I came across a monitoring product known as Kale. http://codeascraft.com/2013/06/11/introducing-kale/
I have not yet evaluated the product. It is open systems software. The ideas that I like about it, however, increase the usefulness of gathering performance statistics.
Kale has two parts that 1) detect anomalies, and 2) look for other instances matching that anomaly. This data mining of the monitored data was quite interesting to me. It appears that the framework could be used or modified to detect and report issues as they develop, rather than after they have occurred.
I wonder how much of that concept will become a reality. Share your thoughts by adding comments below, or send an Email to btaylor@sswug.org.
Cheers,
Ben
$$SWYNK$$
Featured White Paper(s)
Extending the Value of SCOM with Spotlight on SQL Server
read more)
Featured Script
dba3_sysStatistics_sysDBA
not a script just an explanation of the purpose of two databases often used on SS2k and earlier DBMS’…. (read more)