Ted, You want to use connect time to figure out MB/sec on FICON? Be my guest. I would rather beat my head a against a steel spike :-)
The busier any component in the path gets, the more inaccurate it becomes. Path includes the MP on the host and the storage, ISL if you have any cascading, command latency, business continuance processing, and multitude of other variables. It's your dog Ted, and I would name him Caveat Emptor. BTW IO Service units never counted blocks. It counted the EXCP provided by the Access Method of which some counted blocks, and some did not. I agree IO connect time was a far more accurate way of counting IO Service Units, and I have used it in just about every IPS I have built. However Connect Time has never been a reliable metric for calculating MB/sec on ESCON. You usually accomplish this by placing average connect time into blksize buckets, and multiplying that blksize by the SSCH Rate. The ratio of connect time to data transferred is not linear on ESCON, so you need a collection of magic numbers, not just one. When I have been forced to use a single multiplier it has always been one that assures an overestimate. (I made my living doing Remote Copy sizing in half a dozen countries for over half a decade. I have several scars and T shirts). Ron > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of > Ted MacNEIL > Sent: Thursday, September 23, 2010 7:00 PM > To: IBM-MAIN@bama.ua.edu > Subject: Re: [IBM-MAIN] Reports for GB per hour to tape > > >The EXCP*BLKSIZE=MB is not true for all access methods. I'd only assume this > is correct for SAM-E, which is probably the majority of tape IO. > > Ron, we've had this argument/discussion before. > > With XA/ESA, or even slightly before, IOC was changed to allow either blocks > or 8.3 ms of connect time. > The latter being more accurate. > > Mind you, connect time has to be multiplied by nominal/actual/ficticious > transfer rate. > > Bogus is as bogus does! (8-{]} > > - > I'm a SuperHero with neither powers, nor motivation! > Kimota! > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO > Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
