I'm not an expert on controllers and drivers. It reachs the limit of what I probably can effectively google. So I wouldn't know it a commit to the buffer of a hard drive would be considered by the zfs layer as a "secure" write when it received a COMMIT request from the NFS layer.
I'll do some testing when I've got some hardware to try back up some of my analysis. At the moment I'm just trying to get a handle on a simple methodology that might be of practical use when reviewing different specs and deciding how large a stripe is required in order to sustain performance requirements. Nicholas On Thu, Apr 23, 2009 at 2:38 AM, Todd Pisek <[email protected]> wrote: > Nicholas Lee wrote: > >> If we turn this around - with 32KB NFS blocks in order to get 90MB you >> need about 2900 iops or .35 ms path latency. Network latency (*) on 32KB is >> at least .28ms. So .07 ms is left for the sync to disk. I doubt this is >> possible with rotational disk - even with a RAID0 array. According to the >> numbers I have I'd say it is possible with a strip of eight X25-M SSD drives >> or maybe 4 STEC Zeus SSD drives. >> > Is the iSCSI command queue depth > 1? If so, the disk write of command N > can overlap with the network transfer of command N+1. As long as the disk > xfer rep rate holds, over time the bandwidth approaches effective > interconnect bandwidth. >
_______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
