More Performance and Capacity Predictability out of RMF and SMF
Project and Program:
MVS,
MVS Core Technologies
Tags:
Proceedings,
2018,
SHARE St. Louis 2018
The good news is that SMF and RMF/CMF are the most robust machine-generated operations data in the data center. And the richer the data, the more predictive and prescriptive the analytics can be if they properly mine the data for that content.
The problem is in how hard it is to correlate and understand all the data and what it really means for performance and cost-efficiency of your workloads on your specific z/OS infrastructure configuration.
There is so much data, and it requires so much expertise to interpret it, that it is virtually impossible to do proactively if you are relying on eyeballing hundreds of static reports to do it.
To advance the state of the art, the computer has to do more of the work for the human analyst by using z/OS specific Artificial Intelligence techniques.
This session explores new approaches to analyzing the operational data and in particular how predictability can be dramatically affected with an additional type of artificial intelligence automatically applied to the data.
This modernized approach will be discussed from a vendor-neutral perspective. Some specific z/OS performance problem examples will be shown.-Brent Phillips-Intellimagic
Back to Proceedings File Library