Micron Abandons 3D XPoint Memory Technologyby Billy Tallis on March 16, 2021 6:00 PM EST
In a sudden but perhaps not too surprising announcement, Micron has stated that they are ceasing all R&D of 3D XPoint memory technology. Intel and Micron co-developed 3D XPoint memory, revealed in 2015 as a non-volatile memory technology with higher performance and endurance than NAND flash memory.
Intel has been responsible for almost all of the commercial volume of 3D XPoint-based products, under their Optane brand for both NVMe SSDs and persistent memory modules in the DIMM form factor. Micron in 2016 announced their QuantX brand for 3D XPoint products, but never shipped anything under that brand. Their first and only real product based on 3D XPoint was the X100 high-end enterprise SSD which saw very limited release to close partners. Micron has now decided that further work to commercialize 3D XPoint memory isn't worth the investment.
Micron currently owns the only fab equipped to mass-produce 3D XPoint memory: the Lehi, Utah fab that was formerly the home of the Intel-Micron flash and 3D XPoint joint venture IMFT. Intel and Micron began splitting up their partnership in 2018, first parting ways for 3D NAND flash memory development, followed by dissolving the 3D XPoint partnership after completing development on the second generation 3D XPoint. In 2019, Micron exercised their rights to buy out Intel's share of the IMFT fab, leaving Micron as the sole owner of the fab and Intel in the position of buying 3D XPoint wafers from Micron to use in Optane products. Intel's Optane products have not been enough to fully utilize the capacity of that fab, and Micron's non-GAAP operating profits have been taking a hit of over $400 million per year in underutilization charges.
Micron is now putting that 3D XPoint fab up for sale, and is currently engaged in discussions with several potential buyers. Intel is the most obvious potential buyer, having recently begun the long process of selling their NAND flash and flash-based SSD business to SK hynix while keeping their Optane products. Intel has already moved their 3D XPoint R&D to Rio Rancho, NM but has not built up any 3D XPoint mass production capacity of their own; buying the Lehi, UT fab would save them the trouble of equipping eg. their NAND fab in Dalian, China to also manufacture 3D XPoint.
However, Intel is not guaranteed to be the buyer of the Lehi, UT fab. They've doubtless had opportunities to do so before as Intel and Micron unwound their partnership. Micron states that the Lehi, UT fab could be used to produce analog or logic ICs, not just memory—and that converting it to large-scale manufacturing of DRAM or NAND flash memory would not be as appealing to Micron as simply expanding capacity at their other existing fabs. With widespread semiconductor shortages affecting almost all corners of the industry, this fab is likely to sell quickly even if the buyer needs to put substantial effort into retooling.
Micron does not have a direct replacement lined up for 3D XPoint memory technology, but continues R&D into new memory and storage technologies. Micron's announcement is emphasizing a pivot toward developing memory products that will use the Compute Express Link (CXL) interface, which promises to be a vendor-neutral interface for DRAM and non-volatile memories such as 3D XPoint.
Post Your CommentPlease log in or sign up to comment.
View All Comments
gdansk - Thursday, March 18, 2021 - linkReally strange. The reason I don't buy Optane is that it is too expensive. Yet they are not making as much of it as they could. Doesn't it seem like a price vs. demand problem? Lower price = more demand. But clearly Micron isn't expecting an increase in orders from Intel or they wouldn't put it up for sale.
melgross - Thursday, March 18, 2021 - linkIt’s always possible that it’s just too difficult to produce, and so can’t be brought down significantly in cost. If that’s true, I can understand Micron’s decision. Intel might feel differently, but then the question is why advances haven’t come regularly, as they do with DRAM and NAND?
I’ve seen a lot of new technologies that were supposed to change everything die, because they couldn’t be produced at the prices they needed to be at. This could be one of those. Theoretically feasible, but not feasible financially, or even technologically.
edzieba - Friday, March 19, 2021 - linkMicrons repeated mention of CXL as an alternative to 3DXP is... odd, to say the least. It would be like Seagate announcing they are abandoning HAMR in favour of PCIe-based drives: the storage hardware itself and the interface that talks to it are two different things. There is nothing whatsoever stopping Micron creating CXL-connected devices that use 3DXP other than Micron themselves.
Billy Tallis - Saturday, March 20, 2021 - linkYeah, it's a bit of a non sequitur at first glance. But I think the underlying admission is that 3DXP and any similar persistent memory cannot be successful at scale without a pre-existing ecosystem around an interface like CXL. Relying on Intel to add 3DXP support directly to their CPUs and letting them have a monopoly on Optane DCPMM prevented 3DXP from being commoditized in the way it needed to be to be successful. I expect Micron doesn't want to try again with a different memory technology unless it has a clear path to a large TAM.
JoeDuarte - Saturday, March 20, 2021 - linkDisappointing. I very much want to see Instant Computing. We should've gotten there by 2000 or so, but instead PCs and phones are inexplicably slow. Everything but video encoding should be instant, every app should be open and ready in 100 ms or so. Booting should be a second or two tops. Anything happening in an app should also be at no more than 100 ms latency.
Optane would've been good for Instant Computing, though we probably shouldn't need it. The latency of an SSD over PCIe 3.0 or 4.0 should enable those 100 ms latencies easy. There must be an enormous number of CPU cycles spinning just to do simple things like opening an app, things that were actually faster in the 1980s. Microsoft and Intel could do very well with a new generation of OSes and architectures, but they have no vision.
AnnonymousCoward - Monday, March 22, 2021 - linkExactly. I do not understand why Microsoft doesn't care about user experience. The latest version of Office takes about 5 seconds to open up Word, and there's often lag in right-click menus and other pop-ups. Yet Office 2002 is lightning fast. It seems software devs care more about packing in features while not giving a crap about responsiveness.
FunBunny2 - Friday, March 26, 2021 - link"Yet Office 2002 is lightning fast. It seems software devs care more about packing in features while not giving a crap about responsiveness."
way back when (when? no longer recall), Gates answered complaints about Windows (again, which version?) performance with... "let the hardware fix it". that's less and less likely these days.
AnnonymousCoward - Friday, March 26, 2021 - linkThe amount of time I spend each day waiting for files to open probably adds up to minutes. That's on quite modern 2018 hardware (i5, SSD). Xpoint/Optane wouldn't make it any better, either, as you can see from benchmarks that measure real-world performance.
peevee - Friday, April 2, 2021 - link"I very much want to see Instant Computing. We should've gotten there by 2000 or so, but instead PCs and phones are inexplicably slow. Everything but video encoding should be instant, every app should be open and ready in 100 ms or so. Booting should be a second or two tops"
peevee - Friday, April 2, 2021 - linkMBA mismanagement strikes again.