Monday, January 28, 2019

Five Thoughts from 2019 SNIA Persistent Memory Summit

Last week I attended the Persistent Memory Summit in Santa Clara. This is a great one day conference each year bringing together experts on Persistent Memory examples, system support, and applications. The presentations are posted and there is a video as well (thank you SNIA!)

5 thoughts:
  1. Now that persistent memory has moved from a “wouldn’t it be great if we had this?” concept to a “we have some options, now what?” debate…. We need to define “persistent memory” based on the new reality. Rob Peglar and Stephen Bates reminded us that using the term SCM is not politically correct and can only be used in a safe space away miles from a SNIA conference (Starbucks Milpitas worked for me). This is good since it was way too vague and theoretical. Andy Rudoff offered a simple definition: it needs to be address in load/store like memory (Not blocks and pages), and persistent. Speed is in the eye of the beholder but a year ago there was a definition of <2us latency in applications which I liked. The NVDIMM-N and NVDIMM-P definitions would indicate that it does not need to be one type of memory but is a DIMM or system. These simple definitions would seem to eliminate some products that are often referred to as “persistent memory” (side discussion)
  2. The most common persistent memory today arguably is NVDIMM-N which provides us with up to 32GB DIMMS that can be written to like DRAM but never lose data. The challenges here are that the use of DRAM for entire capacity plus NAND plus energy support leads to a high cost that is 3x or more per bit compared to DRAM. As a result, a small amount of systems (typically SANs) use them today. Multiple providers were at the conference and you can buy this persistent memory whenever you wish.
  3. Frank Hady presented Intel Optane Persistent Memory and the applications. Two modes, one is persistent memory (App Direct) and one is memory Mode (which loses data on power cycle). Memory mode is great for adding tons of memory that is somewhat slower and cheaper but it is not persistent per Intel documentation. This is poised to grow rapidly with Intel backing but it is off to a slow start. From talking to customers, most say they still can’t get Optane PM to build their own system and the availability today is running apps on cloud systems. I have details on modes and projected revenue in other publications
  4. NVDIMM-P is proposed as an open source version similar to Optane PM where the architecture supports some DRAM plus NAND or other memory type to optimize for cost. This will allow DIMMs that are LESS expensive than DRAM, higher density, and more non-proprietary options. We need this ASAP! When can I get one???
  5. From the conference, it feels like Infrastructure support and application drivers are ahead of the actual hardware…. This is probably not totally true but there is drive from Intel and SNIA to get all the support in place and the OS supports it and we have applications. Once Intel ships significant volume and competitors start shipping their versions of PM, we can test out all the applications

See more info on our blogs or website. Thanks to Chris Mellor of Register fame for republishing some of my FMS work on persistent memory and Optane with all the gory details and numbers.

Mark Webb

Tuesday, January 22, 2019

Relative Cost and Price for Optane and other Memory

Jan 2019 UPDATED: What are the relative costs for the memory types in new SSDs and DIMMs? We have the estimates below!

This was shown about a year ago and is still a good representation of what is going on. All of the costs are lower (we publish a monthly report with details).... but the summary is still true.

  • 3D XPoint is lower cost than DRAM today and is selling for half the price of DRAM or less in DIMMs and SSDs (volumes are low and yes Intel loses money on this). 
  • 3D XPoint will decrease in cost with ramp and maturity. It is not fully ramped yet as demand is not there yet. Also 2nd Gen will be 30% cheaper.
  • Optane DIMMs are now known to have a controller and DRAM on each DIMM so the DIMM has some additional cost compared to DRAM. Plus they are overprovisioned more than DRAM DIMMS
  • Fast NAND (Low Latency NAND) is still much cheaper and is useful in Fast SSDs and NVDIMM-P applications. We expect Low latency NAND from Samsung, Toshiba, WDC, and Hynix in 2019  (YMTC someday ... and yes we have data on when someday is)

Actual values, assumptions, prices and how these will change over the next two years is available as well

We will be at persistent memory summit if you want to discuss details more. Text us to chat.


Mark Webb

Wednesday, January 16, 2019

Intel Optane H10 Vs Samsung 970EVO

Intel announced an Optane H10 Hybrid SSD at CES last week. How might it compare to Samsung 970EVO on cost and performance?

Intel announced a Optane H10 M.2 SSD which will be in laptops sometime in Q2 or Q3. It combines Intel optane memory (the 16G or 32G cache for HDD) and a Intel QLC SSD all on one module. Optane/3D XPoint acts as a cache for the QLC SSD. By merging them together, Intel sells both NAND and 3D Xpoint and saves the PC manufacturer a M.2 slot. It is two SSDs on one board managed by Intel RST driver and custom firmware

Performance: we can expect that the Optane will perform like other Optane Memory tests. Lower latency on anything cached, slow down to SSD (QLC in this case) performance on huge serial transfers. This alone will make it a high performance SSD. Most likely faster than 970EVO on QD1 latency and speed, slightly slower than 970 on large file transfers or un-cached as QLC is slower than TLC in most applications. Power is unknown but historically, Optane is a power hungry technology

Cost is the key as always: Our data indicates H10 cost is slightly higher than 970EVO at 500GB, slightly lower than 970EVO at 1TB. See details below. Since Intel is highly motivated to sell this product to move its built up NAND and 3D XPoint inventory, they can price it aggressively... even at cost. This should make it a competitive product for PC OEMs. Not HDD cheap, and not as cheap as Intel's pure QLC NVMe SSD but a less expensive SSD with high NVMe performance.

It should be a solid competitor to 970EVO for the highest performance notebook storage.

H10 Review,38387.html

Optane Memory Review,5032.html

970EVO Review,5573.html

We have costs for all SSDs, NAND, and new NVM technologies along with performance numbers for Optane/3d XPoint

Mark Webb