Silicon Motion at CES: 3D NAND support for SM2246EN and roadmap update

2015 was a great year for SSD controller designer Silicon Motion. Their SM2246EN controller was at the heart of some of the best mainstream and value SATA SSDs, while their DRAM-less SM2246XT and their TLC-compatible SM2256 each had several design wins for even more affordable SSDs. At CES, Silicon Motion showed off their full range of products and shared some of their plans to stay competitive through 2016.

The most important development for the SSD market in 2016 will almost certainly be the availability of 3D NAND from companies other than Samsung (who’s been shipping 3D NAND since 2014 and will be rolling out their third generation of it this year). Silicon Motion has updated their firmware for the SM2246EN controller to support 3D MLC NAND, and they showed off drives using 3D NAND flash sourced from Intel, Micron, and Hynix. This demonstrates that Silicon Motion is ready for the transition to 3D NAND and that we can expect drives to be hitting the shelves as soon as the flash itself is available in bulk on the open market. It’s also nice to have independent confirmation that both IMFT and Hynix are on track with their 3D NAND development. Conspicuously absent from the lineup was 3D NAND from the Toshiba and SanDisk joint venture. We already expected them to be last to ship 3D NAND due to their fab for it not being scheduled to begin mass production until this year, so it’s no surprise if they’re keeping things under wraps for a little longer.

To support 3D TLC NAND, Silicon Motion will be releasing a SM2258 controller as the successor to SM2256, but this new controller was not on display and we don’t have information on what other changes it may bring to the table. SM2258 should be ready by the middle of the year, so it shouldn’t be too long before we have more details.

The last big update concerns the SM2260 PCIe SSD controller. A launch date hasn’t been announced, but we were told to expect a more interesting demo at Flash Memory Summit, suggesting it will be ready to ship in the second half of 2016. The expected performance specifications have changed slightly from what we last heard in June 2015: sequential read speed is up from 2200 MB/s to 2400 MB/s while sequential write is down from 1100 MB/s to 1000 MB/s. Random read and write ratings remain at 200K and 125K IOPS respectively. With the exception of random write those numbers are a bit below what Samsung advertises for the 950 Pro, but close enough that SM2260-based drives can probably be competitive by just undercutting Samsung’s pricing by a little bit. 3D NAND support has also been added to the feature list, and NVMe version 1.2 will be supported. To make use of the higher speeds of the PCIe 3.0 x4 interface, the SM2260 uses a dual-core ARM processor instead of the single-core ARC processor used by Silicon Motion’s SATA SSD controllers and the SM2260 has 8 NAND interface channels compared to 4 channels for SM2246EN and SM2256.

Silicon Motion still has no direct successor planned for the DRAM-less SM2246XT controller but they confirmed that all of their controllers could be used in a DRAM-less configuration with appropriate firmware, so a DRAM-less TLC SSD could be built using SM2256 if somebody thought the cost savings were worth the firmware development efforts. Silicon Motion was also showing off their current lineup of solutions for USB flash drives including Type-C and Lightning port support, as well as their eMMC and single-package SSD products intended mainly for industrial, automotive and other embedded applications.

Gallery: Silicon Motion at CES 2016

Marvell Implements Host Memory Buffer for DRAM-less 88NV1140 SSD Controller

The first version of the Non-Volatile Memory Express (NVMe) standard was ratified almost five years ago, but its development didn’t stop there. While SSD controller manufacturers have been hard at work implementing NVMe in more and more products, the protocol itself has acquired new features. Most of them are optional and most are intended for enterprise scenarios like virtualization and multi-path I/O, but one feature introduced in the NVMe 1.2 revision has been picked up by a controller that will likely see use in the consumer space.

The Host Memory Buffer (HMB) feature in NVMe 1.2 allows a drive to request exclusive access to a portion of the host system’s RAM for the drive’s private use. This kind of capability has been around forever in the GPU space under names like HyperMemory and TurboCache, where it served a similar purpose: to reduce or eliminate the dedicated RAM that needs to be included on peripheral devices.

Modern high-performance SSD controllers use a significant amount of RAM, and typically we see a ratio of 1GB of RAM for every 1TB of flash. The controllers are usually conservative about using that RAM as a cache for user data (to limit the damage of a sudden power loss) and instead it is used to store the organizational metadata necessary for the controller to keep track of what data is stored where on the flash chips. The goal is that when the drive recieves a read or write request, it can determine which flash memory location needs to be accessed based on a much quicker lookup in the controller’s DRAM, and the drive doesn’t need to update the metadata copy stored on the flash after every single write operation is completed. For fast consistent performance, the data structures are chosen to minimize the amount of computation and number of RAM lookups required at the expense of requiring more RAM.

At the low end of the SSD market, recent controller configurations have chosen instead to cut costs by not including any external DRAM. There are combined savings of die size and pin count for the controller in this configuration, as well as reduced PCB complexity for the drive and eliminating the DRAM chip from the bill of materials, which can add up to a competitive advantage in the product segments where performance is a secondary concern and every cent counts. Silicon Motion’s DRAM-less SM2246XT controller has stolen some market share from their own already cheap SM2246EN, and in the TLC space almost everybody is moving toward DRAM-less options.

The downside is that without ample RAM, it is much harder for SSDs to offer high performance. Even with clever firmware, DRAM-less SSDs can cope surprisingly well with just the on-chip buffers, but they are still at a disadvantage. That’s where the Host Memory Buffer feature comes in. With only two NAND channels on the 88NV1140, it probably can’t saturate the PCIe 3.0 x1 link under even the best circumstances, so there will be bandwidth to spare for other transfers with the host system. PCIe transactions and host DRAM accesses are measured in tens or hundreds of nanoseconds compared to tens of microseconds for reading from flash, so it’s clear that a Host Memory Buffer can be fast enough to be useful for a low-end drive.

The trick then is to figure out how to get the most out of a Host Memory Buffer, while remaining prepared to operate in DRAM-less mode if the host’s NVMe driver doesn’t support HMB or if the host decides it can’t spare the RAM. SSD suppliers are universally tight-lipped about the algorithms used in their firmware and Marvell controllers are usually paired with custom or third-party licensed firmware anyways, so we can only speculate about how a HMB will be used with this new 88NV1140 controller. Furthermore, the requirement of driver support on the host side means this feature will likely be used in embedded platforms long before it finds its way into retail SSDs, and this particular Marvell controller may never show up in a standalone drive. But in a few years time it might be standard for low-end SSDs to borrow a bit of your system’s RAM. This becomes less of a concern as we move through successive platforms having access to more DRAM per module in a standard system.

Patriot Memory Enters PCIe Storage Market with Hellfire SSDs

Patriot Memory has been selling solid-state drives for about eight years now. To date, virtually all of Patriot’s SSDs have used the Serial ATA interface, which became a performance-limiting factor in the recent years. At the Consumer Electronic…