New Texas Memory Systems patent enhances reliability and improves MTBF with RamSan-70

HOUSTON, USA: Texas Memory Systems Inc. (TMS) sets a new standard for Flash reliability with the introduction of its patented Variable Stripe RAID (VSR) error mitigation technology on the RamSan-70 PCIe Flash card. Failure of a plane (small subsection of a Flash chip) is the second most common kind of Flash failure after Flash block failures.

TMS VSR technology (U.S. Patent #7,856,528) addresses plane and chip failures by dynamically bypassing failed Flash, reducing maintenance requirements and extending operating lifecycles. VSR technology is available exclusively from TMS, the leader in enterprise solid state reliability.

Variable Stripe RAID (VSR) forms a core portion of the TMS SSD reliability technology suite developed through extensive Flash device testing in labs and the field. TMS lab facilities include a burn-in room used to stress test systems for one month prior to shipment. The intense testing process has provided a treasure trove of real-world reliability data for several generations of Flash media. Flash media naturally degrades after hundreds of thousands of write cycles.

TMS has seen three basic categories of Flash media failures: block failures, disabling a tiny part of a chip; plane failures, affecting a small part of a chip; and complete chip failures. Block failures are typically handled by error correcting code (ECC). VSR is an improved management approach for plane and chip failures.

Most enterprise SSDs use a simple type of error mitigation called RAID which groups Flash media into “stripes” containing equal numbers of chips. RAID technology can reconstruct data from a failed Flash chip. Typically, when a chip or part of a chip fails, the RAID algorithm uses a spare chip as a virtual replacement for the broken chip. But once the SSD is out of spare chips, it needs to be replaced. VSR technology allows the number of chips to vary among stripes, so bad chips can simply be bypassed using a smaller stripe size.

Additionally, VSR provides greater stripe size granularity, so a stripe could exclude a small part of a chip rather than having to exclude an entire chip if only part of it failed. With VSR technology, TMS Series-7 products like the RamSan-70 require less maintenance than average SSD designs.

According to Dan Scheel, president of Texas Memory Systems, “Variable Stripe RAID technology is another TMS innovation that maximizes return on SSD investments and minimizes maintenance downtime.”

To further understand the benefits of VSR, consider a hypothetical SSD made up of 25 individual Flash chips. If a plane failure occurs that disables an eighth of one chip, a traditional RAID system would remove a full 4 percent of the raw Flash capacity. TMS VSR technology bypasses the failure and only reduces the raw Flash capacity by 0.5 percent, an 8x improvement. TMS tests show that plane failures are the second most common kind of Flash device failures, so it is very important to be able to handle them without wasting working Flash.