Design Article
Not All MLC SSDs are created equal
Scott Stetzer, STEC, Inc.
1/21/2012 3:05 PM EST
Improving MLC SSD Endurance
To effectively improve MLC SSD endurance, advanced media management techniques need to be implemented within the SSD controller (and managed internal to the NAND flash memory). Using digital signal processing (DSP) functionality, the SSD controller can support complex management technologies such as dynamically adjusting the program and erase charges to achieve higher levels of efficiency. This enables algorithms to not only monitor how cells are being used but also predicts how reads and writes will affect performance over the long term. In this way, the algorithm can spread out cell wear to ensure consistent performance over time.
The performance, cost efficiency, and reliability of an SSD ultimately come down to the intelligence of its SSD controller. Extending the life of NAND flash memory in an SSD requires a flexible and intelligent controller architecture that can analyze and adapt dynamically to the changing characteristics of the flash cells as they age. Unmanaged MLC-based SSDs are not suitable for the continuous duty cycles and heavy workloads required for today’s demanding network environments. Through the use of SSDs built with adaptive flash management algorithms and advanced signal processing techniques, IT professionals can confidently accelerate network access with near-zero downtime and advanced data protection at the lowest cost per I/O. Extending the useful life of an enterprise MLC flash-based SSD delivers a higher return on the SSD investment.
New Technology
To slow the breakdown of the floating gate electrons and to flatten MLC flash reliability and endurance concerns, technology exists today that is capable of extending the life of the NAND flash memory within an MLC SSD. At STEC, we have developed CellCare™ Technology which is based on four generations of proprietary SSD controller architecture. CellCare Technology uses a combination of write and erase management techniques, read level adjustments, write softening techniques, DSP methods for signal/bit detection, and other management technologies (as discussed earlier) to increase NAND cell life and deliver endurance benefits.
Since SSDs slow down and degrade due to the increased use of error correcting codes and retries required to overcome read issues (the more retries required, the slower the performance over the life of the drive), CellCare Technology-enabled MLC SSDs deliver fewer errors and retries reducing drive degradation while improving performance and extending its useful life (see Table 2).
Table 2 provides a quick comparison of the NAND flash memory types with MLC CellCare Technology added improves SSD endurance:
The SSD controller transforms raw, generic flash into an enterprise-class SSD by dynamically measuring and managing flash media wear to extend SSD life without sacrificing performance. It uses advanced signal processing functions to improve flash endurance, adaptive flash management algorithms to actively manage NAND wear, and media error management and tuning functions that adjust over time and the life of the media. It also integrates other enabling technologies, as outlined in Figure 1.
In addition to CellCare Technology, STEC’s intelligent SSD controller includes:
Figure 2 shows enterprise SSD endurance based on the total drive writes per day for the NAND flash memory types based on the 32 nanometer technology node. So, the real question again is can an SSD manufacturer guarantee up to 30 full capacity writes per day for 5 years using MLC media? The answer is ‘yes’ as outlined in Figure 2.
Summary
Selecting an MLC flash-based SSD for enterprise deployment requires more than just an evaluation of write and read performance. High-performance, enterprise-class SSDs have to be able to balance performance and endurance to provide sustained and deterministic network behavior as drives age, while providing twice the density and cost-effective price points versus SLC flash-based SSDs.
The performance, cost efficiency, and reliability of an MLC flash-based SSD ultimately comes down to the intelligence of its SSD controller. Extending the life of NAND flash requires a flexible and advanced controller architecture that can analyze and adapt dynamically to the changing characteristics of flash cells as they age. Unmanaged MLC-based SSDs are not suitable for the continuous duty cycles and heavy workloads required for today’s demanding network environments. Through the use of MLC SSDs built with adaptive flash management algorithms and advanced signal processing techniques, like STEC’s CellCare Technology, systems equipment manufacturers, integrators and data center managers can confidently accelerate data access with near-zero downtime and advanced data protection at the lowest cost per I/O – reducing the total cost of ownership of MLC flash-based SSDs.
If you liked this article...
About the Author
With more than twenty years of experience in storage product marketing and leadership, Scott Stetzer is Vice President of Technical Marketing for Enterprise Storage at STEC and is responsible for bringing the company’s enterprise storage strategy to the industry. He has extensive storage experience in field applications engineering, software development, and technical marketing with such global companies that include Western Digital, Maxtor and Quantum. Mr. Stetzer is regarded as a futurist of storage drive technologies and has delivered keynote speeches and presentations on a variety of storage topics at key industry events that included the International Disk Forum (Japan), Storage Network World (Europe), the Data Storage Forum (China) and Flash Memory Summit (US).
To effectively improve MLC SSD endurance, advanced media management techniques need to be implemented within the SSD controller (and managed internal to the NAND flash memory). Using digital signal processing (DSP) functionality, the SSD controller can support complex management technologies such as dynamically adjusting the program and erase charges to achieve higher levels of efficiency. This enables algorithms to not only monitor how cells are being used but also predicts how reads and writes will affect performance over the long term. In this way, the algorithm can spread out cell wear to ensure consistent performance over time.
The performance, cost efficiency, and reliability of an SSD ultimately come down to the intelligence of its SSD controller. Extending the life of NAND flash memory in an SSD requires a flexible and intelligent controller architecture that can analyze and adapt dynamically to the changing characteristics of the flash cells as they age. Unmanaged MLC-based SSDs are not suitable for the continuous duty cycles and heavy workloads required for today’s demanding network environments. Through the use of SSDs built with adaptive flash management algorithms and advanced signal processing techniques, IT professionals can confidently accelerate network access with near-zero downtime and advanced data protection at the lowest cost per I/O. Extending the useful life of an enterprise MLC flash-based SSD delivers a higher return on the SSD investment.
New Technology
To slow the breakdown of the floating gate electrons and to flatten MLC flash reliability and endurance concerns, technology exists today that is capable of extending the life of the NAND flash memory within an MLC SSD. At STEC, we have developed CellCare™ Technology which is based on four generations of proprietary SSD controller architecture. CellCare Technology uses a combination of write and erase management techniques, read level adjustments, write softening techniques, DSP methods for signal/bit detection, and other management technologies (as discussed earlier) to increase NAND cell life and deliver endurance benefits.
Since SSDs slow down and degrade due to the increased use of error correcting codes and retries required to overcome read issues (the more retries required, the slower the performance over the life of the drive), CellCare Technology-enabled MLC SSDs deliver fewer errors and retries reducing drive degradation while improving performance and extending its useful life (see Table 2).
Table 2 provides a quick comparison of the NAND flash memory types with MLC CellCare Technology added improves SSD endurance:
The SSD controller transforms raw, generic flash into an enterprise-class SSD by dynamically measuring and managing flash media wear to extend SSD life without sacrificing performance. It uses advanced signal processing functions to improve flash endurance, adaptive flash management algorithms to actively manage NAND wear, and media error management and tuning functions that adjust over time and the life of the media. It also integrates other enabling technologies, as outlined in Figure 1.
In addition to CellCare Technology, STEC’s intelligent SSD controller includes:
- Secure Array of Flash Elements™ (S.A.F.E.) Technology that improves reliability by providing mechanisms to recover from NAND flash page-, block-, die- and chip-failures while maximizing the Mean Time Between Failure (MTBF) and the Mean Time To Data Loss (MTTDL).
- PowerSafe™ Technology which provides the data persistence and instant data backup and recovery protection in the event of an unplanned power failure.
- Data path protection to protect critical data from corruption.
- Predictive read optimization that minimizes loss of performance over the useful life of the drive.
- Advanced ECC to measure and manage flash media wear.
Figure 2 shows enterprise SSD endurance based on the total drive writes per day for the NAND flash memory types based on the 32 nanometer technology node. So, the real question again is can an SSD manufacturer guarantee up to 30 full capacity writes per day for 5 years using MLC media? The answer is ‘yes’ as outlined in Figure 2.
Summary
Selecting an MLC flash-based SSD for enterprise deployment requires more than just an evaluation of write and read performance. High-performance, enterprise-class SSDs have to be able to balance performance and endurance to provide sustained and deterministic network behavior as drives age, while providing twice the density and cost-effective price points versus SLC flash-based SSDs.
The performance, cost efficiency, and reliability of an MLC flash-based SSD ultimately comes down to the intelligence of its SSD controller. Extending the life of NAND flash requires a flexible and advanced controller architecture that can analyze and adapt dynamically to the changing characteristics of flash cells as they age. Unmanaged MLC-based SSDs are not suitable for the continuous duty cycles and heavy workloads required for today’s demanding network environments. Through the use of MLC SSDs built with adaptive flash management algorithms and advanced signal processing techniques, like STEC’s CellCare Technology, systems equipment manufacturers, integrators and data center managers can confidently accelerate data access with near-zero downtime and advanced data protection at the lowest cost per I/O – reducing the total cost of ownership of MLC flash-based SSDs.
If you liked this article...
- Head to the Memory Designline homepage for the latest updates in memory and storage.
- Sign up for the Memory Designline Newsletter, 2X a month delivered to your mailbox with the latest highlights from the site.
About the Author
With more than twenty years of experience in storage product marketing and leadership, Scott Stetzer is Vice President of Technical Marketing for Enterprise Storage at STEC and is responsible for bringing the company’s enterprise storage strategy to the industry. He has extensive storage experience in field applications engineering, software development, and technical marketing with such global companies that include Western Digital, Maxtor and Quantum. Mr. Stetzer is regarded as a futurist of storage drive technologies and has delivered keynote speeches and presentations on a variety of storage topics at key industry events that included the International Disk Forum (Japan), Storage Network World (Europe), the Data Storage Forum (China) and Flash Memory Summit (US).
|
|
|
|
|
Navigate to related information
sharps_eng
1/22/2012 3:36 PM EST
This technology illustrates the pressure to create workarounds for the recent Moore's Law crunch that means smaller geometries are not appearing fast enough to meet demand.
Previous EDC/ECC and other flash-'nursing' initiatives failed because bigger chips appeared that allowed the protection to be implemented at a higher level, in software. Hardware was only necessary for custom high-integrity applications.
STEC have a window of opportunity to make MLC work for a wider range of applications before a memory breakthrough pushes the density up again cheaply enough to compete. But is that breakthrough in sight? I personally love FRAM but can it be made dense enough? I think not.
Production will also only be available when a big fab becomes surplus to DRAM or flash requirements. No-one will build a fab for FRAM speculatively, I think.
Perhaps a slowdown will create spare FAB capacity?
Sign in to Reply
DrWattsOn
2/5/2012 1:08 PM EST
Agreed about FRAM: love the idea, but I doubt density will reach levels high enough for use in computers as Storage: maybe BIOS/EUFI.
I like the materials from STEC, but not able to find any products identified as containing their technology, even searching all their links. Looks like vaporware to me.
Sign in to Reply
markhahn
11/15/2012 12:04 PM EST
where did this figure of 30 full-device writes per day come from? I'm sure there's a market for that, but it has to be fairly small. obviously, most storage and computation is more consumer-like, with read-mostly loads, and often much sparser duty cycles than 24x7. it's easy to find very cheap SSDs today that peak at 500 MB/s and 80k iops and still offer 3-5 year warranties. commodity storage is cheap enough to simply use above-device redundancy to solve issues of reliability and permanence.
STEC's pitch seems to be pretty intensive engineering at the device level - laudable, but do people buy these inherently more expensive (and apparently slower) devices and trust them without any above-device redundancy (raid, etc)?
Sign in to Reply