It appears the end may be in sight for Intel’s beleaguered Optane memory business. Tucked inside a brutal Q2’2022 earnings release for the company (more on that a bit later today) is a very curious statement in a section talking about non-GAAP adjustments: In Q2 2022, we began the divestment of our Intel Optane memory business. Intel’s earnings report also notes that the company is taking a $559 million charge for “Optane inventory impairment” this quarter.
Beyond these two items, there is no further information about Optane in Intel’s results release or their accompanying presentation card. We have contacted the company’s representatives to seek more information, and are awaiting a response.
Taking these elements at face value, it appears that Intel is preparing to shut down its Optane memory business and development of associated 3D XPoint technology. To be sure, there is a high degree of nuance here around the Optane name and product lines here ̵[ads1]1; which is why we are looking for clarification from Intel – as Intel has several Optane products, including “Optane memory” “Optane persistent memory” and “Optane SSDs”. Nevertheless, in Intel’s previous earnings announcements and other financial documents, the entire Optane business unit has traditionally been referred to as its “Optane Memory business”, so it appears that Intel is actually discontinuing the Optane business unit, and not just the Optane Memory product .
Update: 6:40 p.m. ET
Following our request, Intel has issued a brief statement regarding the Optane discontinuation. While it doesn’t provide much in the way of further details about Intel’s exit, it does confirm that Intel is indeed exiting the entire Optane business.
We continue to rationalize our portfolio in support of our IDM 2.0 strategy. This includes evaluating divesting businesses that are either not sufficiently profitable or are not core to our strategic objectives. After careful consideration, Intel plans to discontinue future product development in the Optane business. We are committed to supporting Optane customers through the transition.
Intel, for its part, used 3D XPoint as the basis for two product layouts. For its data center customers, it offered Optane Persistent Memory, which packaged 3D XPoint into DIMMs as a partial replacement for traditional DRAMs. Optane DIMMs offered greater bit density than DRAM and combined with its persistent, non-volatile nature made it an interesting proposition for systems that needed massive working memory sets and could take advantage of its non-volatile nature, such as database servers. Meanwhile, Intel also used 3D XPoint as the basis for several storage products, including high-performance SSDs for the server and client markets, and as a smaller high-speed buffer for use with slower NAND SSDs.
However, 3D XPoint’s unique characteristics have also been a challenge for Intel since the technology was launched. Despite being designed for scalability via layer stacking, manufacturing costs for 3D XPoint have remained higher than NAND on a per-bit basis, making the technology significantly more expensive than even higher-performance SSDs. Meanwhile, Optane DIMMs, while filling a unique niche, were just as expensive and offered slower transfer speeds than DRAM. So despite Intel’s efforts to offer a product that can cross the two product areas, for workloads that don’t take advantage of the technology’s unique capabilities, 3D XPoint ended up being neither as good as DRAM or NAND at their respective tasks—making Optane products a hard sell.
As a result, Intel has lost money on its Optane business for most (if not all) of its lifetime, including hundreds of millions of dollars in 2020. Intel doesn’t break out Optane revenue information on a regular basis, but on one – on occasions where they have published these figures, they have been well in the red on an operating income basis. In addition, reports from Blocks & Files have claimed that Intel is sitting on a significant oversupply of 3D XPoint chips – on the order of two years’ worth of stock as of earlier this year. All of this underscores the difficulty Intel has faced in selling Optane products, increasing the cost of a write-down/write-off, as Intel is doing today with its $559M Optane write-down charge.
Consequently, a potential spin-off for Optane /3D XPoint has been in the tea leaves for some time now, and Intel has taken steps to change or limit the business. Most notably, the dissolution of the Intel/Micron IMFT joint venture left Micron in possession of the sole manufacturing plant for 3D XPoint, while Micron abandoned its own 3D XPoint plans. And after producing 3D XPoint memory into 2021, Micron eventually sold the factory to Texas Instruments for other uses. Since then, Intel hasn’t had access to a high-volume factory for 3D XPoint – but if the inventory reports are true, they haven’t needed to produce more of the memory for a long time.
Meanwhile on the product side of things, the discontinuation of the Optane business follows Intel’s earlier withdrawal from the client storage market. Although the company has released two generations of Optane products for the data center market, it has never released a second generation of consumer products (eg Optane 905P). And after selling its NAND business to SK Hynix (which now operates as Solidigm), Intel no longer manufactures other types of client storage. So retiring the remaining data center products is the logical next step, albeit an unfortunate one.
Intel’s previous Optane Persistent Memory Roadmap: What Will Never Be
Overall, Intel has chosen to wind down the Optane/3D XPoint business at a critical time for the company. With their Sapphire Rapids Xeon CPUs launching this year, Intel was previously slated to launch a third generation of Optane products, most notably their “Crow Pass” 3rd generation persistent DIMMs, which would, among other things, update the Optane DIMM technology to use a DDR5 interface. While development of Crow Pass is presumably complete or nearly complete at this point (given Intel’s development schedule and Sapphire Rapids delays), actually launching and supporting the product will still incur significant upfront and long-term costs, requiring Intel to support the technology for another generation.
Instead of Optane persistent memory, Intel’s official strategy is to turn to CXL memory technology (CXL.mem), which allows volatile and non-volatile memory to be connected to a CPU over a CXL-compatible PCIe bus. This would achieve many of the same goals as Optane (non-volatile memory, large capacity) without the cost of developing an entirely separate memory technology. Sapphire Rapids, on the other hand, will be Intel’s first CPU to support CXL, and the overall technology has much wider industry support.
AsteraLabs: CXL memory topology
Still, Intel’s retirement of Optane/3D XPoint marks an unfortunate end to an interesting product line. 3D XPoint DIMMs were a novel idea even if they didn’t quite work, and 3D XPoint made for ridiculously fast SSDs thanks to its huge random I/O advantage – and that’s a feature it doesn’t look like any other SSD -suppliers. will be able to fully replicate anytime soon. So for the solid state storage market this marks the end of an era.