An nvram device would have to do at least 14k iops at 32kB packet size to hit .07ms latency. The iRAM supposedly can do about 3k iops at 32kB [1] and the uMem device (which looks interesting) [2] is supposedly 3.5x as fast. So it's possible.
Shame there is no readily available nvram PCIe device that "just works" for solaris. [1] http://www.xbitlabs.com/articles/storage/display/ssd-iram_5.html [2] http://www.mail-archive.com/[email protected]/msg19297.html On Wed, Apr 22, 2009 at 1:25 PM, Bob Friesenhahn < [email protected]> wrote: > On Wed, 22 Apr 2009, 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. >> > > I am using a RAID array with a non-volatile (battery backed) write cache. > > Was this number with nfs+zfs or ufs? I thought zfs was more strict with >> NFS >> and did not allow the commit to memory NFS3+ version of sync NFS. >> > > This is with nfs+zfs. > > Bob > -- > Bob Friesenhahn > > [email protected], http://www.simplesystems.org/users/bfriesen/ > GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ >
_______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
