Xiotech’s acquisition of Seagate’s Advanced Storage Architecture (ASA) group brought substantial intellectual property and some impressive management credentials (e.g. Steve Sicola, Ellen Lary, Richie Lary). While Xiotech has not been overly aggressive in marketing the magic of its intellectual property, the company has filed 75+ patents in connection with this technology, and enough information is now available to assess its performance in the field.
A key aspect of Xiotech’s IP is the Intelligent Storage Element (ISE). ISE is a brick-like unit that houses Datapacs holding ten 3.5” disk drives or twenty 2.5” disk drives each that are offset in a manner to reduce vibration and heat (disk drive killers) and increase the reliability of storage systems. The problem Seagate gurus set out to solve relates to the fact that disk drives were never designed for vertical insertion into storage arrays, a technique used by many OEM’s to pack more density into an array. This approach puts stress on drives and failures become more common. Seagate of course was in a unique position to address this issue and minimize failure rates as well as address the significant problem of false positives, which, it says, cause more than half the field failures.
The team at Seagate re-examined the design of disk drives and made substantial firmware changes and also altered how drives were mounted in arrays. The process took the better half of the decade but resulted in more stable and reliable drives that could perform self-monitoring and autonomic healing.
Field Results
The obvious question is how reliable are these systems as compared with conventional approaches? Xiotech has provided relevant data:
ISE is the underlying technology that goes into Emprise, the brand name for Xiotech’s storage arrays, and more than 1,600 ISE’s are in the field. Typically, an ISE has two Datapacs installed, so there are more than 3,200 Datapacs in the field. Xiotech measures reliability at the Datapac level,which is equivalent to 10 ordinary disk drives in a bay. In other words, Xiotech has data on more than 32,000 disk drives, or about the equivalent of 2,000 ordinary 16-slot drive bays.
Xiotech began shipping ISEs and Datapacs in June 2008, giving it 16 months of data. Here are the results:
- No data stored on ISE has ever been lost in the field due to Datapac sparing level going below zero.
- In only two cases has a Datapac reached a percent sparing level that called for a replacement. These replacements were done for $0 under Xiotech’s five-year hardware warranty policy.
- Xiotech has of course seen other failures of heads, drives going off line, false positives, etc., but the dynamic reliability management capabilities of ISE have mitigated those failures in the field.
- Xiotech estimates customers have experienced the equivalent of four drives failing (as compared with normal un-intelligent drive bays)—i.e. four un-repairable drive failures—since June of 2008.
- Annual failure rates (AFR) for disk drives according to research conducted by Schroeder and Gibson at Carnegie Mellon, range from 2-13%, significantly higher than vendor spec sheets imply (i.e. AFR’s of less than 1%).
- This means Xiotech customers, who have experienced the equivalent of four drive failures in 16 months, are seeing more than two orders of magnitude better reliability than conventional disk devices, which in the same time period would have seen several hundred failures using the Schoeder, Gibson figures.
Does this Matter?
You bet. With the popularity of RAID in the 1990’s, users have seen dramatic reduction in data loss. However the enormous number of disk drive spindles installed in the field means that despite high MTBF’s, drives fail frequently and need to be replaced. This can mean incremental cost and headaches to customers in the form of having to maintain spare invent