On 29/11/2017 10:08 AM, Anthony Thompson wrote:
What about Spark on Z? IBM are giving it away.
I haven't tried it, but from the documentation and presentations it
seems pretty heavyweight for simple SMF reporting.
I see a couple of other issues:
- IBM were saying how great it was at parallelizing work, however that
is a bit of a red flag for me. Parallelizing is great if you have e.g. a
multi core workstation where you hit 100% usage on one core while the
others are idle. That isn't usually the case on z/OS. Most z/OS
customers would prefer speeding up by reducing total CPU usage than
spreading it across multiple CPUs. It would be fairly uncommon on z/OS
for this type of job to run at a priority where single CPU capacity is
the limiting factor.
- I think the reporting is SQL based? SMF data was not designed to be
queried using SQL. I have been down that path. On the other hand, Java
classes are beautiful for working with SMF data (not quite as nice as C#
but still very good).
--
Andrew Rowley
Black Hill Software
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
