The SSD market has been divided—with multi-level cell (MLC) NAND memory the technology of choice for client applications, such as laptops, while single-level cell (SLC) NAND has been preferred for enterprise applications.
Recently enterprise MLC technology has advanced to where it can successfully manage challenging enterprise applications—at a cost far lower than SLC—as I discussed in my article, “How to Know if Your SSD will Go the Distance.” As a result, IDC predicts that while MLC accounted for only 27% of the enterprise SSD revenue in 2010, MLC revenue will rise to 52% in 2012 and will hit 60% in 2013.
This blog post describes why enterprise MLC is indeed the best SSD choice for increasing numbers of enterprise applications due to its combination of performance, endurance/reliability, support—and low cost.
Performance Today’s MLC SSDs offer enterprise class performance in terms of boot speed, random IOPS and other critical measures. The average enterprise class SSD can push 48,000 IOPS for random reads and 15,000 IOPs for random writes, significantly more than enterprise class hard disk drives that achieve 350 to 400 random read/write IOPS.
New benchmark tests allow organizations to determine whether a particular SSD delivers enterprise-class performance. These tests include the Storage Performance Council (SPC)-1C storage benchmarks as well as the Solid State Storage Performance Test Specifications (SSS PTS) from the Storage Networking Industry Association (SNIA) Solid State Storage Technical Work Group and the SNIA Solid State Storage Initiative (SSSI).
Endurance/Reliability Endurance/reliability was once the major barrier to MLC use in the enterprise. Memory cells on an SSD drive have a finite write/erase life before they wear out and the drives become unreliable. 3-bit-per-cell MLC NAND flash can be erased only 100 to 500 times before it goes bad and standard 2-bit-per-cell MLC NAND can withstand only 3,000 to 5,000 Program/Erase (P/E) cycles. In contrast, maximum write/erasures for SLC NAND are 100,000, which made them more reliable.
Newer enterprise MLC technologies, however, are incorporating improvements that enhance endurance and reliability. Advanced media management techniques, such as wear leveling algorithms that dynamically allocate blocks across the entire SSD, help ensure that write activity is spread evenly across the flash, reducing wear and tear on individual cells and extending the useful life of the SSD. Enterprise-class MLC NAND is available with additional screening, trimming and testing to increase reliability and endurance. In addition, MLC interface controllers now provide end-to-end data protection, such as extra Input Output Error Detection and Correction Code and support for the ANSI T10 DIF standard (Protection Information), to ensure high reliability.
New endurance standards--JEDEC JESD218A and JESD219 from the JEDEC Solid State Technology Association--enable enterprises to confirm that SSDs meet enterprise endurance standards.
Support While most MLC SSD vendors have little enterprise experience, several leading vendors that understand the needs of large enterprises now offer enterprise MLC products. These companies understand the criticality of enterprise data and avoid unintentionally modifying or losing data during the storage process. They can help with proof-of-concept and architectural design for the development of complex solutions within the data center infrastructure and meet all quality and reliability requirements. They can also deliver the technical resources necessary to support the complex and expensive qualification process and have a global support and distribution infrastructure.
Cost While enterprise MLC technology offers the capabilities necessary to support enterprise applications, they cost far less than SLC technologies that were once preferred in the enterprise.2 The popularity of enterprise MLC will likely continue to grow as MLC increasingly addresses enterprise needs for performance, reliability and support—while offering the lowest costs per gigabyte.
About the Author Teresa brings over 20 years IT Product Strategy, Marketing and Sales experience to her role as Senior Product Marketing Manager, for Enterprise Products at Seagate. Teresa is responsible for messaging and positioning the Pulsar, Savvio, and Cheetah enterprise product families and well as Seagate’s Self-Encrypting Drive security technology. Prior to joining Seagate Teresa enjoyed work at industry leaders such as Platinum Technologies, and Sterling Commerce. Teresa received her MBA from the University of Colorado – Boulder and her BBA in Management Information Systems from Texas A&M University.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.