I'm running randread and randwrites with fio. I would expect something out of the cache device- instead nothing.....
I've considered using bcache, but I would have to build that for centos 7. Not out of the question, just not in my immediate wheelhouse. Thanks for the response, very much. Zac On Monday, September 19, 2016 at 4:41:45 PM UTC-4, quadstor wrote: > > You could try btrace on the device to see the IO activity. > > If you are running tests with zeroed data, then they will be > deduplicated. So there will be negiligible data written to the SSD > disk by the vdisk. > > On Mon, Sep 19, 2016 at 4:33 AM, Zac Durham <zacd...@gmail.com > <javascript:>> wrote: > > There's a few topics out there on this subject but I can't get my head > > around any of them.... > > > > I have 3 x 2TB in software (mdadm) RAID5 (yeah, yeah I know) and a spare > > 120G SSD to use as cache. Currently I use flashcache to back the entire > md > > device but do not realize any benefit in doing so (nor do I see any > activity > > on the cache device when performing I/O against the array). > > > > Where is my approach flawed? I've also tried fronting the qs vdisks > > themselves with the SSD device to no avail either. > > > > Any help would be appreciated. > > > > -Z > > > > > > -- > > You received this message because you are subscribed to the Google > Groups > > "QUADStor Storage Virtualization" group. > > To unsubscribe from this group and stop receiving emails from it, send > an > > email to quadstor-vir...@googlegroups.com <javascript:>. > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "QUADStor Storage Virtualization" group. To unsubscribe from this group and stop receiving emails from it, send an email to quadstor-virt+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
