l...@garlic.com (Anne & Lynn Wheeler) writes: > FICON initially just mapped on top of underlying fibre channel > ... ignoring all the i/o program batching ... finally FICON starts > with zHPF and TCW.
re: http://www.garlic.com/~lynn/2012m.html#2 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee http://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee http://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee As I've referenced numeruous times (ever since early 90s fibre channel standards meetings) ... FICON was effectively really ugly and unnatural layering of traditional mainframe channel paradigm on top of base fibre channel ... enormously cutting effective thruput. zHPF and TCW significantly improves FICON thruput ... partially recreating the original underlying fibre channel paradigm as another layer ontop, starting to approach the underlying, native fibre channel thruput. individual zHPF fibre channel peak at 92k IOPS (significant improvement compared to base FICON but less than 1/10th 1M IOPS emulex fibre channel for e5-2600) and 104 zHPF channels has theoritical peak of 9.6M IOPS ... but max z196 peaks 2M IOPS and the 14 system assist processors saturates (all running at 100% utilization) at theoritical peak 2.2M IOPS. -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
