Hello Erik, Friday, June 23, 2006, 2:35:30 AM, you wrote:
ET> So, basically, the problem boils down to those with Xeons, a few ET> single-socket P4s, and some of this-year's Pentium Ds. Granted, this ET> makes up most of the x86 server market. So, yes, it _would_ be nice to ET> be able to dump a tuning parameter into /etc/system to fix the cache ET> starvation (and other related <4GB RAM) problems. However, I have to ET> say that working with PAE is messy, and, honestly, 64-bit enabled 1U/3U ET> servers are dirt cheap now. So, while I empathize with the market that ET> has severe purchasing constraints, I think it's entirely reasonable to ET> be up front about needing a 64-bit processor for ZFS, _if_ we've ET> explored expanding the 32-bit environment, and discovered it was too ET> expensive (in resources required) to fix. It's reasonable. However in many datacenters with x86 gear there're a lot of old 32bit Xeon boxes and in many cases ZFS will be useless (too slow). Most of these boxes have 2GB/4GB of ram and ZFS will use less than 512MB of its caches (metadata/data). I know I can switch to x64 server - but it doesn't make sense for that application - and we won't just throw away all old hardware. Right now we're testing UFS (it's running Linux in a production and we want to switch to Solaris). -- Best regards, Robert mailto:[EMAIL PROTECTED] http://milek.blogspot.com _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
