I woke up today with a nice blog post from Tom’s Hardware titled, “Intel Kills Optane Memory Business Entirely“. While this may not e a surprise to some, it sure jolted me. As I started to look into this, there’s more to this story.
No More Optane Memory “In the Future”
Although the blog title makes it seem like Intel Optane memory is history, a closer look reveals a bit more. As the Tom’s Hardware blog mentions, Intel has stated “Intel plans to cease future product development within its Optane business.” Note, it did not say, Optane memory is no longer available, NOR did it say it won’t be available in the upcoming Sapphire Rapids launch next year. No, it said “future.” So I wouldn’t dismiss Optane memory just yet.
Supply Could be an Issue
The bigger concern, is the supply. Since I don’t have visibility into their supply chain, it’s all speculation Intel has 3 generations of Optane Memory (Series 100, 200 and 300.) I believe the 100 series, which was part of the Cascade Lake generation of blade servers, likely has a very low level of supply. However, the Ice Lake generation of blade servers uses the 200 series Optane Memory and it’s likely to have a normal supply. IF Intel does deliver the 300 series Optane Memory on Sapphire Rapids blade servers, I would imagine the supply would be spotty. Regardless, the supply will definitely inhibit the architectural designs that could have used Optane.
For the past 8 or so years, I’ve been talking with people about how storage class memory could improve application performance since it’s a lower latency device compared to NVMe. I also talk about Compute Express Link, or CXL, which is a cache coherent design that will allow memory expansion beyond the box. CXL is still a few years away, but (apparently) poses enough of a threat to Intel’s CEO, Pat Gelsinger, that it was one of the reasons he cited for getting out of the Optane business. To be honest, I don’t know that the potential performance of CXL is near the same as you get with Intel Optane memory but it must be close enough.
Another alternative that is available near-term, is an offering from Liqid that simply put, takes NVMe devices and presents them to the host O/S as memory. Again, I don’t know the performance compared to Optane DIMMs, but it is another angle worth pursuing for organizations looking for something other than Optane memory.
What are your thoughts about this announcement from Intel to discontinue Optane memory? What do you think the future holds for replacing it? Let me know in the comments below, and as always – thanks for reading!
Kevin Houston is the founder of BladesMadeSimple.com. With over 24 years of experience in the x86 server marketplace Kevin has a vast array of competitive x86 server knowledge and certifications as well as an in-depth understanding of VMware virtualization. He has worked at Dell Technologies since August 2011 and is a Principal Engineer supporting the East Enterprise Region and is also a CTO Ambassador for the Office of the CTO at Dell Technologies.
Disclaimer: The views presented in this blog are personal views and may or may not reflect any of the contributors’ employer’s positions. Furthermore, the content is not reviewed, approved or published by any employer. No compensation has been provided for any part of this blog.