Intel to shrink Optane memory work

It looks like the end may be imminent for Intel’s embattled Optane memory business. Tucked inside a brutal release of the company’s Q2’2022 earnings (more on that later today) is an intriguing statement in a section that talks about non-GAAP adjustments: In the second quarter of 2022, we began closing the Intel Optane memory business. Additionally, Intel’s earnings report also notes that the company is charging $559 million in “optane inventory impairment” fees this quarter.

Other than these two items, there is no other information about Optane within the Intel earnings release or associated presentation suite. We have reached out to company representatives for more information, and are awaiting a response.

Taking these items at face value, it appears that Intel is preparing to close the Optane memory business and develop the associated 3D XPoint technology. To be sure, there is a high degree of nuance here about the Optane name and product lines here – which is why we’re looking for clarification from Intel – as Intel has several Optane products, including “Optane Memory” and “Optane Persistent Memory”. Optane SSDs. However, in Intel’s previous earnings releases and other financial documents, the entire Optane business unit was referred to as their “Optane Memory business,” so it appears that Intel is actually shutting down the Optane business unit, not just the Optane Memory product.

Modernization: 6:40 PM ET

At our request, Intel sent a short statement on Optane Filtering. While it doesn’t offer much in the way of more details about Intel’s exit, it does confirm that Intel is indeed exiting the entire Optane business.

We continue to rationalize our portfolio to support our IDM 2.0 strategy. This includes evaluating the liquidation of companies that are either not profitable enough or are not essential to our strategic objectives. After careful consideration, Intel plans to discontinue future product development within the Optane business. We are committed to supporting Optane customers during the transition process.

Intel, in turn, used the 3D XPoint system as the basis for two sets of products. For its data center customers, it offered Optane Persistent Memory, which bundled 3D XPoint into DIMMs as a partial alternative to traditional DRAM. Optane DIMMs offered greater bit density than DRAM, and combined with their continuous and non-volatile nature to make an interesting proposition for systems that need huge working memory pools and can take advantage of their non-volatile nature, such as database servers. Meanwhile, Intel has also used 3D XPoint as the basis for many storage products, including high-performance SSDs for the server and customer market, and as a smaller, high-speed cache for use with slower NAND SSDs.

The unique features of 3D XPoint have also been a challenge for Intel since the technology was launched. Although designed for scalability via layer stacking, 3D XPoint’s manufacturing costs continued to rise above NAND on a per-bit basis, making this technology much more expensive than higher-performance SSDs. Meanwhile, Optane DIMMs, while filling a unique niche, were equally expensive and offered slower transfer rates than DRAM. So, despite Intel’s efforts to deliver a product that could transcend the producer space, for workloads that don’t take advantage of the technology’s unique capabilities, 3D XPoint ended up not being nearly as good as DRAM or NAND at their tasks — making Optane products difficult Sale.

As a result, Intel has been losing money in its Optane business for most (if not all) of its life, including hundreds of millions of dollars in 2020. Intel does not disclose Optane revenue information on a regular basis, but on the first level – on occasions that Where they posted these numbers, they were well in the red on an operating income basis. In addition, reports from Blocks & Files have claimed that Intel is experiencing a significant oversupply of 3D XPoint chips — within two years of inventory as of earlier this year. All of this underscores the difficulty Intel has had in selling Optane products, and add to the cost of delisting/delisting, which is what Intel is doing today with Optane’s $559 million impairment charge.

Thus, the potential decline of Optane/3D XPoint has been in the tea leaves for a while now, and Intel is taking steps to alter or scale back the business. Notably, the Intel/Micron IMFT joint venture solution left Micron owning the sole production plant for 3D XPoint, while Micron abandoned its own 3D XPoint plans. After producing 3D XPoint memory through 2021, Micron finally sold the original product to Texas Instruments for other uses. Since then, Intel has not been able to access mass storage for 3D XPoint – although if the stock reports are correct, it won’t need to produce more memory in some time.

Meanwhile, on the product side, the termination of Optane’s business follows Intel’s previous withdrawal from the customer storage market. While the company has released two generations of Optane products for the data center market, it has never released a second generation of consumer products (such as the Optane 905P). After selling their NAND business to SK Hynix (now Solidigm), Intel no longer produced other types of client storage. So cutting out the remaining data center products is the next logical, albeit unfortunate, step.

Intel’s Previous Stable Optane Memory Roadmap: What Never Will

Overall, Intel chose to terminate the Optane/3D XPoint business at a critical juncture for the company. With the launch of Sapphire Rapids Xeon CPUs this year, Intel was previously scheduled to release its third generation of Optane products, chief among them the “Crow Pass” 3research and development Create continuous DIMMs, which among other things will update Optane DIMM technology to use the DDR5 interface. While Crow Pass development is assumed to be complete or nearly complete at this point (due to Intel’s development schedule and Sapphire Rapids delays), in fact product launch and support 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 pivot toward CXL memory technology (CXL.mem), which allows volatile and non-volatile memory to be attached to the CPU via a CXL-capable PCIe bus. This would achieve many of the same goals as Optane (non-volatile, large-capacity memory) without the costs of developing a completely discrete memory technology. Sapphire Rapids, in turn, will be the first CPU from Intel to support CXL, and the overall technology has much broader industry support.

AsteraLabs: a CXL memory topology

However, Intel’s retirement of Optane/3D XPoint marks the unfortunate end of an interesting product lineup. 3D XPoint DIMMs were a novel idea even if it didn’t quite work, and 3D XPoint was made for ridiculously fast SSDs thanks to its massive random I/O feature – a feature that looks like no other SSD vendor will be able to fully replicate anytime soon . So for the solid state storage market, this marks the end of an era.

Leave a Reply

%d bloggers like this: