> > John, > > > > I have been told that CFW must be turned off for Hyperswap, but I never > > understood why. You say that CFW must be written to disk for Hyperswap, > > but > > why can all the other writes still be deferred in cache/NVS? > > > > Ron > > > > > > > > Hyperswap would be one of those circumstances were you would also want > > > the > > > I/O to be written to the disk. In that case you must ensure that CFW > > is > > > not > > > being used. > > > > > > John > > > > > G'day Ron, > > sorry, I hadn't seen your update. > > 'written to the disk' wasn't exactly what I meant to say! > > 'All writes have to be in the secondary disk subsystem before the I/O is > completed' is more accurate. This is not the case for CFW. > > John
G'day Ron, I should know better than to argue with you! You're right. It looks like the GDPS group also misunderstood CFW. Excerpt from DS8000 introduction and planning. cache fast write A form of the fast-write operation in which the storage server writes the data directly to cache, where it is available for later destaging. Excerpt from GDPS You should eliminate any known exploitation of Cache Fast Write (CFW). Disk write operations using CFW (Cache Fast Write) are written into cache but not to the disk. Having CFW operations in progress at the time a HyperSwap occurs can yield unpredictable results since there is no corresponding mirrored cache content in the secondary disk subsystem. Any known exploiters of CFW (such as DFSORT) should be customized to not use CFW. And just to complete the thread. To turn CFW on or off at the LCU level use IDCAMS SETCACHE. --- John ---------------------------------------------------------------------- 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
