Optane, We Hardly Knew Yeh – EEJournal – EE Journal
Buried in Intel’s recent and somewhat dismal second-quarter 2022 financial results was a line item under “Non-GAAP adjustment or measure” that read: “Optane inventory impairment” for the sum of $559 million. Ouch. About 60 percent of the way through the particular “Forward-Looking Statements” section of typically the financial earnings statement is the phrase “the wind-down of our Intel Optane memory business. ” Bye, bye Optane.
This is a CEO-level decision, and it can’t have been an easy one for Pat Gelsinger. Optane Persistent Memory has been a foundational cornerstone associated with Intel’s differentiation strategy for its Xeon server CPUs plus Core CPUs for PCs and laptops, as you can see from this short, uplifting video:
Intel’s relationship with Optane non-volatile memory has not been a quick fling. Optane started as a joint project between Intel in addition to Micron a decade ago but was formally announced in July 2015. Back then, it was called 3D XPoint by Micron and even Optane simply by Intel. Although the technical together with material science details underlying Optane memory were never fully explained, it’s generally considered to be a phase-change type of resistive RAM (ReRAM).
Optane non-volatile memory was available in two forms: SSDs (SATA and NVMe) and “Persistent Memory” DIMMs that could plug into existing DIMM slots, if the system’s memory controller understood Optane DIMMs’ unique timing requirements, which differed from SDRAM timing needs. Only Intel CPUs and some Intel FPGAs incorporated memory controllers that will understood Optane Persistent Memory DIMMs, which was both the competitive advantage for Intel and fairly of a new hindrance in order to wider use of Optane Persistent Memory space DIMMs.
Optane has always been caught between DRAM and NAND Flash memory space. DRAM is a bit faster and additionally somewhat more expensive on some sort of per-bit basis than Optane memory, not to mention NAND Flash memory is much slower and much less expensive on a per-bit basis. Optane Persistent Storage DIMMs also offered more bit density than SYNCHRONOUS DYNAMIC RANDOM ACCESS MEMORY DIMMs. The latest (and presumably last) Optane selection guide says of which Intel offered non-volatile Optane Persistent Recollection DIMMS in 128, 256, and 512 GByte capacities.
Although a few companies adopted Optane Persistent Memory, the benefits of Optane technology never really sold most customers, and the envisioned volume sales needed to drive down manufacturing costs didn’t materialize. Worse, storage analyst and also consultant Tom Coughlin believes that “… Intel has been subsidizing its Optane products in order to keep the price low enough to generate demand (roughly half the price of MASS, in $/GB). ”
The End regarding Optane
Micron announced the fact that it would cease development of THREE DIMENSIONAL XPoint in March, 2021. Further, this company said that it has been selling often the lone 3 DIMENSIONAL Xpoint semiconductor fab within Lehi, Utah to Texas Instruments, which meant that Intel would have to be able to find another fab for you to host Optane production in case it wanted to continue selling Optane products. With Optane’s unique materials and processing requirements, I can just hear the managers of Intel’s processor fabs yelling, “Not in my back yard! ”
Intel’s idea of using Optane Prolonged Memory to help differentiate the processors isn’t a bad one, but clearly your idea wasn’t boosting CPU sales. Q2 2022 quarterly revenue for the Client Computing Group (that’s PC as well as laptop CPUs) was down 25% versus Q2 2021 while sales for the exact Datacenter and AI Group (that’s where Xeon lives) were down 16%. This is hardly Optane’s fault. The year 2021 was a booming year for Intel, thanks to accelerated purchases of Personal computers, laptops, plus servers caused by COVID-related changes to the way people work. Last year, every semiconductor maker including Intel was running fabs around the clock and shipping every chip that tested good. It was a great year with regard to chip makers. This year was bound to be much less exciting in terms of growth regarding semiconductor manufacturers.
Optane’s termination is not the first time Intel has exited a memory business. Most recently, Intel announced that will it had been selling its NAND Flash company along with their SSD businesses based on NAND Flash in addition to Optane storage in October, 2020, just a few months before Gelsinger took over as Intel’s CEO. SK hynix bought the particular NAND Adobe flash business, and even a new company, Solidigm (an SK hynix subsidiary) took over Intel’s SSD business. Intel entered typically the Flash recollection business inside 1988 along with NOR Flash devices together with later focused on NAND Flash, which usually started marketing in much greater volumes. That’s because NAND Expensive memory, especially multi-level and additionally 3D NAND Flash ram, costs a lot less per bit that NOR Flash devices.
An even bigger storage area divestiture occurred in 1985 when Intel famously leaving the LITTLE business. Robert Noyce not to mention Gordon Moore left Fairchild Semiconductor and also founded Intel specifically to make semiconductor memories. The company literally created the commercial DRAM enterprise with this introduction involving the 1103 DRAM throughout October, 1970, and CORDIAL NIP sales drove the company’s top as well as bottom lines for several years. However , DRAM had been perhaps too successful of a product. Everyone wanted in order to play around the DRAM market. Exactly fifteen years after announcing the world’s first commercial DRAM, Intel announced of which it was exiting the MASS business due to a good glut connected with worldwide LITTLE manufacturing capacity, mostly on Japan, which often eroded CORDIAL NIP pricing and margins to be able to unacceptable levels, at least intended for Intel.
Andy Grove plus Gordon Moore, respectively Intel’s president in addition to CEO at the time, had a famous, private conversation about DRAMs in 85, when Intel’s DRAM market share has been clearly eroding. Grove asked, “What would happen if somebody took us over, got rid of us — what might the new guy do? ” “Get out of the mind business, ” Moore replied.
Grove and even Moore didn’t wait for you to be replaced. They quickly exited often the DRAM organization. Grove later described that decision in your 2012 interview with National Public Radio (NPR) correspondent Laura Sydell: “It seemed to be an emotional decision. We had already been the first to introduce the product and build the business … In retrospect, getting out of DRAMs when we did was the best business decision we ever made. ” After Moore and Grove made basically to terminate its DRAM business, Intel laid off roughly some third of its workforce, more than 7000 employees, and closed multiple manufacturing plants. During that same NPR interview in 2012, Grove said it required ten many years to turn Intel around together with get the company back on track after the fact that fateful choice.
Nearly four decades later on, Intel TOP DOG Pat Gelsinger found himself facing a fabulous similar selection regarding Optane non-volatile remembrance. With your decision to help jettison NAND Flash reminiscence and SSDs already made before his return to Intel, Gelsinger has now ditched Optane as well. No doubt that’s due to the fact he needs a leaner, trimmer company to go up against chief CPU rival AMD.
AMD’s CPUs have become much more competitive together with Intel’s items with respect to both performance and additionally efficiency, so there’s no room in Intel to get money-losing goods. Coincidentally, Intel also put its drone business, Intel Drone Light Shows, on the chopping block, likely for the same reason. (Kimbal Musk, Elon’s brother, bought that business enterprise from Intel. ) At Intel, each employee and every product line is now from battle stations and on high alert using deflector shields set to double front.
The particular Endless Quest for Storage Class Memory
Intel’s Optane was one of many entries – perhaps the exact most successful entry to date : in the quest for “storage-class memory” or even SCM, which in turn attempts in order to deliver low-cost, non-volatile bit storage, like disk drives and SSDs but having greatly improved, DRAM-like access times. Optane largely achieved this goal for the most part, yet apparently not really the “low-cost” aspect, or perhaps at least, not low manufacturing cost. So , the particular pursuit of an SCM technologies that truly challenges SD RAM continues.
There are many candidates pertaining to such a technology, and these candidates have been around meant for a long time. I’ve been writing about SCM designed for at minimum a decade not to mention a half and about alternative non-volatile random access memory technologies just for four decades. As of today, none of these technologies are competitive against SDRAM regarding capacity or maybe cost per bit with data storage space, but that doesn’t prevent aspirational SCM vendors through continuing to be able to develop their non-volatile recall technologies. There’s a pot of gold at typically the end for the SCM rainbow. You simply need to find this right leprechaun.
Personally, I’m rooting for the purpose of MRAM ~ magnetic RAM – not necessarily because I think that MRAM is close, it’s recently been close way too long. MRAM devices have been on often the market for years. So far, they price too much money for each bit and also their device-level capacity will be too lower. No, I am rooting with respect to MRAM since I’d feel a certain closure for you to see your return about magnetic ram memory to it is previous position at the pinnacle of the retention hierarchy.
Im just barely old enough to help remember when Intel announced the 1103 DRAM inside 1970. Within five years of its introduction, DRAM dethroned magnetic-core cognizance, which had reigned for two decades. MIT’s Jay Forrester developed the first successful magnetic-core memory plane and installed it in the groundbreaking MIT Whirlwind computer on August 8, 1953.
Forrester’s magnetic-core memory changed an older electrostatic ability to remember technology that will stored bits in specially built cathode ray tubes (CRTs). Whirlwind’s electrostatic CRT memory experienced a very slow access time in 25 microseconds, many times reduced than what was required. The slow electrostatic storage space bogged down the entire pc and prevented the machine from reaching its performance goals. Typically the installation from magnetic-core secure digital reduced Whirlwind’s memory-access period to 9 microseconds as well as eased of which problem.
Whirlwind’s groundbreaking magnetic cores were installed just one month prior to I was initially born. For the next 20 yrs or so, magnet core usb sat atop the computer storage device Ziggurat, until Intel’s MASS toppled core memory’s rule. The computer industry’s complete conversion from permanent magnet cores to DRAM was initially so swift that many 3rd-party companies making core-memory boards for popular mainframes and minicomputers – companies by using now-unfamiliar names like Ampex, DATARAM, Electronic Memories, plus Fabri-Tek : either switched from using permanent magnetic cores in order to DRAMs as well as vanished seemingly overnight. (In reality, it took about five decades for primary memory to be able to dwindle in addition to die. ) DRAM offers been king of storage space mountain when it comes to 50 several years now.
Many technologies seek to topple the ruler and there are several candidates aspiring to become the one true SCM technological innovation:
- Faster Flash
- Ferroelectric RAM MEMORY
- Resistive RAM
- Phase-change memory space (PCM) ~ Optane was basically in this class
I have written regarding all these candidate SCM technologies over the years. They always seem just over the exact horizon. Like practical fusion power generation, they appear to be perpetually ten years away. Somehow, they all seem for you to be a bit further away now that Optane has fallen by the wayside. Until there’s a breakthrough, which could happen any minute (or not), DRAM continues to reign. Long live the king. So long Optane. All of us hardly knew yeh.