... The rule of thumb I've heard is that you shouldn't run "channel busy + director port busy" over 50%. ...
I've run up to 65-70% with no problem. Which is why I'm asking for evidence. ... I'd say if you can go to 8 channels in width, without incurring an unacceptable cost, go for it. ... We have low profit margins and a flat market. 'Nuff sed! ... It's better redundancy, resiliency, and the need for more data faster only increases over time so extra width is generally a good thing. ... Flat. There used to be an old joke: 3 paths for performance & 1 for redundancy. When the first 3990's came out. I have been in shops where they wasted their money because some 'Storage Guru' said they needed the extra paths because they 'only' had 4. I've presented the same arguments before, asking for evidence: 1. 'High' Channel Busy Delay. 2. 'High' I/O response times. Just because a series of paths are 'busy' doesn't impact response. (By impacy, I mean make it more than measurably worse than times of low usage). My preference is to go to 1.4, and IRD (if/when we get more than one DASD sub-system), and 'float' the channels. Then, if there is evidence, of I/O delay due to channels, upgrade. We're half way through our conversion to 1.4 from 2.10. -teD (The secret to success is sincerity. If you can fake that, you've got it made!) ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
