This is Alex McDonald's Typepad Profile.
Join Typepad and start following Alex McDonald's activity
Join Now!
Already a member? Sign In
Alex McDonald
Recent Activity
Alex McDonald is now following Dave Hitz
May 5, 2010
I never expected to convince you. Or Farley for that matter, but I can safely say he's driving down the wrong path. What we're doing with the technology will just have to come as a surprise in that case!
Toggle Commented Feb 25, 2010 on 2.042: bring out your dead! at the storage anarchist
1 reply
Hi Barry, Alex McDonald aka @alextangent here. Full details of the non-issue over on Marc Farley's blog -- he's an even bigger conspiracy theorist that you! Here's the bottom line; Georgens may be right about storage tiering. But he didn't say it because there's something inherent in WAFL that makes it difficult. Quite the reverse.
Toggle Commented Feb 24, 2010 on 2.042: bring out your dead! at the storage anarchist
1 reply
Very informative piece, Barry. I liked it. On point 1; performance might take a bit of a knock moving to RAID-5 or RAID-6 from RAID-10. You might want to check out RAID-DP here; it doesn't have that downside. On point 4; there's work going on in the SCSI T10 Standards Committee to address some of the issues of "hole punching". I blogged on that and some of the techniques that we (NetApp) use to address the specific problem under Windows; http://blogs.netapp.com/shadeofblue/2008/10/hole-punching-f.html Perhaps you need a 17; the future is virtualized storage. "Real" SANs and "real" FC (just to wind Chuck up a bit here) can't make it into your guarantee. And that other long-time EMC favourite; locality of reference, particularly to do with workloads like Oracle. Poof. Out the window. More like this post please!
1 reply