Micron: We're pulling the plug on 3D XPoint. Anyone in the market for a Utah chip factory? Micron is stopping development of 3D XPoint technology and shifting resources into memory products that use the Compute Express Link (CXL). 3D XPoint is the storage-class memory used in Intel's Optane brand SSDs and Persistent Memory (DIMM) products. CXL is an industry standard interface that enables flexible interconnection …
"Intel told newswires in a statement: "Micron's announcement doesn't change our strategy for Intel Optane or our ability to supply Intel Optane products to our customers.""
Must have a massive stock of unsold stuff (compared with the number of customers) if your manufacturing partner pulling out of the business doesn't affect supply.
Honda just announced that all manufacturing is halted in the U.S. and Canada. Semiconductors are cited as critical.
Can a nand fab make logic parts, at lower densities?
Would any upgrades be cost-effective?
Although Intel itself may soon be conducting a fire sale on logic fans.
When it comes to storage, it seems that despite all the hype, solid state storage will not take over from HDDs. NAND Flash is hurtling fully into QLC and beyond, with increasingly diminishing write endurance and skyrocketing latency, even as HDDs are getting a massive boost in storage density from HAMR and speed (IOPS, access latency & bandwidth) from multi-actuator technology.
Maybe after NAND Flash is well and truly dead can PCM like 3D XPoint begin to take over, but as things stand, it seems that the imminent demise of magnetic mechanical storage has been greatly exaggerated.
>NAND Flash is hurtling fully into QLC and beyond, with increasingly diminishing write endurance and skyrocketing latency
People want cheap, not super-expensive but fast enough to use an x16 bus and huge endurance. Partly Intel's fault, not enough lanes off of consumer CPUs, gamers will favor their GPU every time. The server space will use huge RAM caches backed by battery.
Spinning rust has been a niche product for years, relegated to the lowest of the low end or power users who need to store multiple terabytes of data; for everyone else flash is spacious enough and ten times as fast. It doesn't need to die *completely* to become irrelevant.
What has Intel done right in the last several years?
The only obvious thing, but which in no way is conducive to regaining any and all lost viability, is that they have been reading their own press reports...of ten years ago. And believing those.
The whole point of persistent byte addressable memory is merging memory and storage. Do it right and the persistent memory chipmakers take tens of billions of dollars per year of revenue from the DRAM makers. At the application level, this means user code can write to memory and persist information in the time it takes to do a cache flush, not the time it takes to send a write down the NVMe storage stack.
When the Micron/Intel technology only hit about 10^6 cycles of write endurance, that meant it couldn't be a DRAM replacement as seen from application software, because there had to be wear leveling even more aggressive than flash.
Which meant it was in the I/O path, not the memory path, out of the CPU, and had long latencies compared to DRAM.
It had failed at that point. I salute the Intel people who made a market for 3DXpoint anyhow. But the victory was hollow and the technology should have failed the basic sanity checks before a fab was built.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
