We have unique (off-host) application servers back-ending a mobile app, which sends requests into our CICS ecosystem using unique CICS userids, which is how we separate them out (as there is massive reuse of exactly the same transactions between numerous applications in our environment) for MWP, using summarised 110's based on userid.
On Mon, 24 Jun 2019 at 22:36, Laurence Chiu <[email protected]> wrote: > Hi Andrew > > Thanks for the information. I am making further enquiries about the way our > CPU is charged. The red book is particularly interesting because it seems > to suggest that you can identify mobile workload without having to make any > changes to your applications but use WLM and RMF instead. That might remove > a major barrier to the possibility of proceeding down this path. > > On Sat, Jun 22, 2019, 6:05 AM Andrew Sica <[email protected]> wrote: > > > Hi Laurence, > > > > Jumping in here with a bit of general info: > > > > Mobile Workload Pricing is a way of mitigating the impact from mobile > > requests to the rolling four-hour average. It's a sub-capacity offering > > (and only really makes sense in that space), so you need to have > > implemented sub-cap SW pricing (i.e. AWLC, CMLC, etc.). > > > > Basically, you need to be able to tag and track the CPU time consumed by > > mobile transactions. They need to be isolated from work stemming from > other > > sources and originate on an approved mobile device (a smartphone, for > > example). The redbook I link below discusses the criteria for isolation. > > > > If you can easily do this (and meet any qualification criteria) it may > > very well be a no-brainer. If not easily done, you'd need a deeper > analysis > > to understand what work is necessary to architect the required isolation > > and how much benefit it would actually buy you (i.e. whether and by how > > much it would lower your peaks). > > > > The least amount of work to implement mobile comes with the ability to > tag > > workloads within WLM. Details on that and more can be found here: > > https://www.redbooks.ibm.com/redpapers/pdfs/redp5359.pdf > > > > Here is an updated (2016) US announcement letter for mobile pricing: > > > https://www-01.ibm.com/common/ssi/cgi-bin/ssialias?subtype=ca&infotype=an&supplier=897&letternum=ENUS216-321 > > > > If you have additional questions, please feel free to reach out. > > > > You might also take a look at the options under Tailored Fit Pricing for > > IBM Z, which can provide models that are alternatives to the rolling > > four-hour average. It's worth at least understanding what your options > are > > - for additional info, see: > > https://www.ibm.com/it-infrastructure/z/software/pricing-tailored-fit > > > > > > Take care, > > Andrew Sica > > > > ---------------------------------------------------------------------- > > For IBM-MAIN subscribe / signoff / archive access instructions, > > send email to [email protected] with the message: INFO IBM-MAIN > > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
