"Good example of optimization": Intel eliminates Optane and 3D XPoint

"Good example of optimization": Intel eliminates Optane and 3D XPoint

Intel's quarterly report showed an interesting point not relating to financial performance: in the second quarter, the company started shutting down the development and production of the Optane reservoirs. In addition, the paper stated that the company had to bear the losses of $559 million.

In other words, the company began shutting down Optane, which confirmed Intel's comment to AnandTech's resource: "

As Vice-President Intel Dave Zinsner said: ".

The 3D XPoint memory was announced by Intel in 2015 and the result of a joint project with Mikron was positioned as a solution that incorporated the benefits of operational memory and SSD. The memory supported by the hit address is based on phase transition technology rather than electron capture as in NAND. This allowed 3D XPoint to ensure high reliability of storage devices — about a million re-recording cycles — and high speed of operation, as data do not have to be combined into large blocks.

Using 3D XPoint technology, the company offered two product lines: an energy independent memory in DIMM modules for systems that require large operational memory arrays as well as high-performance storage devices for servers and client machines, and high-speed cache modules for less fast storage devices.

On the other hand, the uniqueness of 3D XPoint has proved to be a major problem for Intel since technology entered the market. Despite the fact that the solution was based on good scalability, the production costs of Optane were higher than those of NAND, so the storage devices were more expensive than the production SSD. On the other hand, the Optane DIM modules at the same price offered more modest speeds than traditional DRAMs. In other words, despite Intel's attempts to combine the benefits of two solutions in the same technology, 3D XPoint proved to be worse than DRAM and too expensive to sell compared to the NAND – Optane.

As a result, Intel simply lost money for most of the life cycle of technology: in 2020 alone, losses exceeded $500 million. The Optane account did not always get into the financial accounts, but when it did, the numbers were always negative. In addition, the company accumulated a significant excess of 3D XPoint chips — at the beginning of this year it was a stock two years ahead, reported the Blocks & Files.

Thus, closing the direction was a matter of time before the decision was made to close the IMLT joint venture owned by Intel and Mikron. The latter was the only producer of 3D XPoint and was subsequently forced to sell Texas Instruments, which later reprofied it. Since then, Intel has lost the production centre of 3D XPoint and apparently no longer needed it. In addition, the company recently sold its Nand business to Korean SK Hynix, which turned it into Solidigm, i.e. Intel took the course of leaving the memory business completely.

The decision to close 3D XPoint was made at a critical time for the manufacturer. With the release of the Sapphire Rapids Xeon chips, the company planned to upgrade the Optane line to the third generation, the most important being the DIMM modules of the Crow Pass family with DDR5 support. They have been either completed or are nearing completion, but the release of the modules will be another burden for Intel: among other things, it will have to provide long-term support to clients.

With Optane leaving, the company will switch priorities to CXL technology that connects the energy-dependent and energy-dependent memory to the processor through the PCIe tire, which is supported by the Sappire Rapids chips. And Intel's failure of 3D XPoint will be a sad finale of an interesting product line. 3D XPoint DIM was undoubtedly an innovative idea, and hardware storage on 3D XPoint could provide productivity that is unlikely to be repeated in the near future. So for the SSD market, this is the end of an entire era.