by Debra Vogler, Senior Technical Editor, Solid State Technology
August 22, 2008 – The relative merits of solid-state drives (SSD) vs. hard-disk drives (HDD) have been discussed by analysts and technologists for some time. Intel’s view on HDDs vs. SSDs came to light during its annual Developer Forum (8/19-8/21, San Francisco, CA), when the company outlined plans for its SATA SSD product family while highlighting the enhanced mobility that SSDs afford notebook computers, as well as the energy cost savings that accrue when they are used in enterprise applications. SST interviewed Troy Winslow, marketing manager of the NAND products groups at Intel, about the announcement.
Several SATA SSD product lines were announced at the event. The multilevel-cell-based (MLC) NAND flash X18-M and X25-M mainstream SATA SSDs for mobile and desktop clients will be available in 80GB and 160GB drives. The 80GB is currently sampling with production slated in the next 30 days; the 160GB version is expected to sample in 4Q08 with production in 1Q09. The single-level cell-based (SLC) NAND flash X25-E Extreme SATA SSD for enterprise, storage, and workstation applications is sampling in 32GB capacity with production slated in the next 90 days, with a 64GB version expected to sample in 4Q08 with production estimated for 1Q09. (More product information is at www.intel.com/go/ssd.)
According to Intel, NAND flash memory has a “license to disrupt HDD,” but to fulfill its destiny, Flash must accommodate the needs of PCs — meeting performance requirements as well as being cost-effective and reliable (see table). Until recently, Winslow told SST, all NAND flash was SLC, i.e., each cell contains one bit of information. With MLC there are two bits/cell of information, effectively doubling the capacity of the device in the same die size. “Because unit costs are based on wafers, now you’re getting double the capacity per wafer, so your costs are cut in half with MLC,” he said.
He noted, though, that “there are tradeoffs when you design with MLC,” one of which is reliability. An SLC device has an order-of-magnitude greater cycling capability. Winslow noted that the standard number of cycles an SLC NAND flash device can attain is 100,000 cycles (a program and erase is one cycle), whereas an MLC device only achieves ~5000-10,000 cycles. “So when you put an MLC device into a computing solution, and if you kept everything constant, the SLC SSD gives you 10× greater write cycling before the device wears out,” he said.
Winslow told SST that Intel was able to achieve an MLC SSD device with high performance and long life. “We can provide at least a five-year useful life in a consumer notebook and that specification far exceeds the user model that we’re benchmarking against,” he said. “We’ve never seen a user write more than 20GB/day every day for five years. Our MLC SSD can allow a user to write 100GB/day every day for five years without wearing out the drive, so we’ve been able to establish that our SSDs — whether SLC or MLC — far exceed the reliability required by the usage models of the end applications.”
Key to achieving performance commensurate with the needs of mobile and PC applications is that the architecture of Intel’s SATA SSDs uses a full 10 channels of flash through the SATA bus, vs. four or six channels used by other SSD manufacturers, Winslow told SST. “So we’ve just expanded the “freeway” of data travel to a full 10 lanes.” But that creates another challenge, he said: how to manage that massive throughput and handle the firmware to respond to the host that is continually asking for a wide variety of read/writes — small, large, sustained, random, and different block sizes, etc.?
The software and firmware used in the Intel SSDs moves the blocks around such that no block gets more worn out than any other — “very similar to rotating the tires on your car,” Winslow explained. “We move the blocks to ensure that all of them are evenly worn to within a 4% delta. It’s a very complicated mechanism in the actual SSD controller to do this.” The other piece of the solution that Intel uses is the principle of write amplification. “We’ve figured out a way to program and erase just what is necessary (in small bytes rather than large blocks) so our drive has tremendous efficiency for the life of the computing environment.”
Also of interest when comparing HDDs to SSDs is what happens to energy costs for data farms. Looking to enterprise applications, Intel estimates that if one rack was filled with 120 SSDs instead of 120 HDDs, the energy savings would stack up as shown in the table below. — D.V.
