1) What is the goal of the environment? 2) How much work needs to be processed? 3) How quickly do you need to get the environment back running after an unplanned outage vs a planned outage. 4) What is the cost for floor space, electricity, heating/cooling? 5) What are the licensing costs? 6) What is the ROI, TCO, - what are your goals?
All platforms have their pros and cons. This does not mean one is superior to another, just different. For zSystems, work horses. You have huge data, a lot of processing power needed and quick responses. And you want to run more than one app at a time. You would not want to use it if you want to just run a little query of a 1000 rows. Excel would probably work fine on a PC. I have some data I would prefer to work in Excel but it is too much for the PC to handle, so I use the mainframe. If you need something that handles IO, huge programs, communicates with almost anything, then look towards zSystems. I worked at a shop that was Huge for zSystems. However, they also ran TPF for a specific application. The reason for TPF is it could IPL quickly. So their appl was not down for very long. With the work done on Parallel Sysplex, DB2 Data Sharing, CICS-Plex, IMS-Plex, the zSystems are up 24x7x365. With advances in SRDF, GDPS, XRC, Gridded Tape Systems, there can be an almost instantaneous roll over. Oh yeah, and there is that z/VM and z/Linux area as well. ;-D I can have a plex and take down an LPAR and my app and users may not even know it happened. I do not necessarily need to move anything, the system will take care of function shipping if it is setup. I would suggest before stating that one environment or another is better than another, try to determine what business problem you are trying to solve and then identify what is best for that business case. zSystems are geared for the future, mobile technology, security, availability, reliability, and I just love 'em. So I am very biased. But not so much not to look to other options when it is a valid business case. Just because one platform does one thing really well, does not make it a solution. Does IBM look at other ways of doing work - yes. Can they get it out quickly - sometimes. Once we can defeat the laws of physics and move huge quantities of data very quickly over very large distances, then I will be much happier. Now I can do it across town. I want to do it across the galaxy. Then the universe. One should always have goals. Just my two cents worth. Lizette > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of IBMZOS > Sent: Thursday, May 28, 2015 5:51 AM > To: [email protected] > Subject: Re: LPAR MOBILITY > > Sorry. i do not know all the solutions, but know what the GDPS and > Hyperswap do. I'm confused to invest many hard work on Parallel Sysplex if > one day, i can do LPM on Z with one clic. I think all the mainframe community > would appreciate... > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
