What about Spark on Z? IBM are giving it away. Ant.
-----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Andrew Rowley Sent: Tuesday, 28 November 2017 9:43 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SAS - DB2 conversion to Java On 24/11/2017 2:50 AM, Barry Merrill wrote: > I'd also be VERY concerned about the execution time if your > applications are high volume under Java versus under SAS. I only know SAS by reputation, but my expectation would be the opposite - I would expect Java to be faster than SAS. I suspect (but haven't been able to measure definitively) that it costs more CPU in TCPIP to FTP the data than to run Java reports on the mainframe - and the Java CPU time is on zIIP. What sort of SMF volumes and processing speeds are typical? I'm currently playing with transaction reporting from CICS monitoring data. On my PC, summarizing data by region and transaction name takes about 45 seconds for around 12 million transactions from compressed CICS data. On the IBM remote development system, about 25 CPU seconds (almost all zIIP) for 2.5 million transactions (I am limited by DASD space so am working on smaller files). I feel like this is fast but would welcome feedback based on real world data. -- Andrew Rowley Black Hill Software ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
