Micron has been moving very quickly for the past couple of years, and that has led to the company’s undisputed leadership in NAND flash layers and DRAM process nodes. The first to ship 176-layer NAND flash chips a year ago, and now 176-layer QLC NAND, Micron packs a huge amount of storage into a single chip. Since it uses a CUA (CMOS Under Array) architecture, the die size is perhaps the smallest in the industry, and the QLC storage augments that to make it more than affordable.
So, what does a company do with a technology like this? Well, in Micron’s case, they have decided to use their advanced NAND flash process technology to squeeze more terabytes of flash into a smaller form factor than anyone has yet done.
Micron has introduced the 2400 SSD, which uses the company’s 176-layer QLC NAND flash to cram 2 Terabytes of storage into a 22x30mm M.2 form factor using only a single-sided PC board. That means that a great big bunch of capacity can fit into even the thinnest of today’s thin and light notebook PCs. That NAND is also available in select Crucial consumer SSDs.
An added plus is that the Micron 2400 SSD uses the PCIe Gen4 interface. This faster protocol takes advantage of the additional speed that is provided by the NAND chip’s charge trap technology and the extra planes made available from the chip’s CUA configuration. (CUA puts the peripheral logic, which sits beside the bit array in most memory chips, below the bits, lopping about 30% off the die size. That means the chip can be made less expensively than alternatives.) Micron tells us that these two technologies, with many others, contribute to deliver a chip that has a 33% higher I/O speed and 24% lower read latency than Micron’s prior generation.
But still, is QLC a good technology for an SSD? Well, IBM certainly thinks it is! At the 2020 Flash Memory Summit, IBM drew a lot of attention by announcing that it was using Micron QLC NAND flash to produce enterprise flash primary storage, a fact that would have seemed impossible only a year or two earlier. As storage experts gain a greater understanding of their workloads, they learn how to tier data so that hot data is pushed to performance storage and colder data is pushed to QLC-based capacity storage. While capacity storage has lower endurance, you get significant justifications for using QLC for a large share of data. These SSDs will allow system architects to assemble systems with a better cost/performance ratio than has been possible in the past.
Micron’s new product is aimed at thin and light notebooks. In that market, size is of the utmost importance, and an HDD can’t compete with an SSD when it comes to packing the greatest number of terabytes into the smallest possible volume. Given that, the Micron 2400 SSD should do very well, which bodes well for the future. I’m looking forward to see what comes next.
