Comments for Permabits and Petabytes

Comment on End of the RAID by Mike Davis

Mike Davis — Tue, 20 Oct 2009 23:08:41 +0000

All good points Jered. I’d say either RAID is dead -or- RAID will be stronger than ever. By this I mean surely some enterprising Marketing Managers will continue to attach a RAID label to their Reed Solomon implementation and co-opt the traditional XOR meaning.

Please be sure to add the point that although configurable M,N protection and parallel restore will allow the market to retain enterprise reliability, one more immediate advantage is the delivery of fail-in-place capabilities…So much of the RAID6 mindshare came out of anecdotes where tech’s replaced the wrong disk in a RAID5 stripe, and we can eliminate that root-cause failure!

FEC-based RAID improves reliability, and facilitates a lower cost-of-service level we’ve been clamoring for for 5 years.

Comment on What do Hash Collisions Really Mean? by jeredfloyd

jeredfloyd — Fri, 19 Jun 2009 16:05:09 +0000

That certainly make it more difficult, but not impossible! You simply need to find input text that generates collisions with both hashes, of which there are still infinitely many.

Comment on What do Hash Collisions Really Mean? by See EYE ay

See EYE ay — Tue, 16 Jun 2009 03:51:50 +0000

Hash Collisions… why don’t you simply sign a messege with two different hashes… This wouldmake it impossible to forge a document.

Comment on Are Fibre Channel and SCSI Drives More Reliable? by JamieIvanov

JamieIvanov — Sun, 15 Mar 2009 13:04:59 +0000

What about JBOD arrays?

This raid issue makes me want to invest in IDE raid heh.

Comment on Dirty Little Secrets About Dirty Little Secrets by scalability.org » Blog Archive » “Top HPC trends” … or are they?

Mon, 23 Feb 2009 12:23:03 +0000

[...] storage) is “breakthrough technology” for archiving. Which is odd. In that industry insiders appear to have a somewhat different opinion on the value of CAS for archiving. Moreover, others point out that CAS is, itself, somewhat of a [...]

Comment on Multiple Drive Failures: RAID 6 vs. RAIN-EC by End of the RAID « Permabits and Petabytes

End of the RAID « Permabits and Petabytes — Tue, 10 Feb 2009 14:39:53 +0000

[...] I explain in an earlier post, the bit error rate of the drives can be catastrophic for RAID. In a RAID 4 or 5 rebuild it is [...]

Comment on Jet Engine on a Duck: You Can’t Retrofit Dedupe by Knut Grunwald

Knut Grunwald — Tue, 03 Feb 2009 16:19:03 +0000

The access time calculated is worst case, since a 16 TB Volume will reside on more than one disk. So you can do checks in parallel. This will accelerate the speed by the number of disks.
But the cost and the power consumption will grow at the same rate …

Comment on Jet Engine on a Duck: You Can’t Retrofit Dedupe by c0t0d0s0.org

c0t0d0s0.org — Tue, 03 Feb 2009 08:44:43 +0000

Some thoughts about deduplication…

Deduplication is one of this big hype topics at the moment. Deduplication sounds good, but … well … hmmm … let´s say it this way: This technology has to mature.

At first there was a link in a internal mailing list. Deduplication is pretty sim…

Comment on Multiple Drive Failures: RAID 6 vs. RAIN-EC by Data Protection’s Black Swan: Seagate Drive Failures « Permabits and Petabytes

Wed, 21 Jan 2009 13:46:50 +0000

[...] I’ve written about before, all data protection schemes depend on statistical failure models. With hard drives, the model allows for two primary types of failure: total spindle failure (i.e. [...]

Comment on The Greening of Storage by The Green Is A Lie « Permabits and Petabytes

The Green Is A Lie « Permabits and Petabytes — Thu, 08 Jan 2009 02:08:48 +0000

[...] They understand the real industry issues around energy efficiency though, which I touched on in “The Greening of Storage” back in November. The real reason enterprise customers, and only enterprise customers, should care about [...]