Wednesday, March 13, 2019

Memory Cycling Capability: Sorry, It's Complex.



I addressed this in a couple presentations at Flash Memory Summit but questions on cycling performance have grown based on 3D Xpoint and QLC. I will highlight why “how many times can you cycle it?” is 10x more complicated that most people realize.










Key items on cycling performance (Random thoughts.... there is a lot more detail)
  • Cycling specifications are based on the criteria used to define success. Rarely do bits fail “dead”. They usually are slow to program or erase, read disturb easily, can’t hold charge for a long time, etc. A device spec could be 800us program time and 100us read time and data retention of 5 years. If I change that to 1200us program, 200uS, and 1 year retention, I could hypothetically increase the cycling capability from 3000 cycles to 10,000 cycles. And what if I changed data retention to 1 hour ???
  • Cycling specs are based on allowed fail rate and die level/system level error correction to deal with it. This is published in detailed bit error rate data (BER).  Basic NAND  allows us to tolerate thousands of  blocks failing to cycle, and millions of bits that sometimes read erratically. I can always increase redundant blocks and add more ECC capability. These “tricks” prevent the end customer from seeing errors. A QLC cell seems less reliable than TLC or MLC but it is possible to make it MORE reliable to the end user with error correction. 
    • In addition to all of the error correction, we can adjust for end user fail rates as well. Is 2% AFR (annual fail rate) OK? 0.7%? 0.2%?  Are you willing to pay 2x the price for the 0.2% fail rate?
  • Based on the above statistics and BER, the size of the array matters. This is very important when we talk about emerging memory. Without error correction, my test chip (a few bits) for MRAM or ReRAM or 3DXP may cycle 10M times. When I put 64K bits in an array it might last 1M times. A 1Gbit array will last 100K times… and then we get into error correction and fail rates.
  • Theory vs Actual. I have seen multiple papers saying that MRAM has infinite endurance or something like 1E12 cycles. But MRAM is a real product now and with that comes real cycling numbers. I don't believe any MRAM product is spec’d with 1E12 cycles. Think 6 order of magnitude less….. best case. There are ways to manage the cycling up and down. The capability starts at 1E12 and then starts to drop when you actually make real devices. This is the problem with universal memory claims because … wait for it … DRAM and SRAM can ACTUALLY last 1E12 cycles in real products. No emerging memory is even close.
  • So, If we just take NAND and things we have seen over the years as examples of the above.
    • NAND theoretically can cycle 1E12… charge trapping is a non-ideality that I will ignore
    • A small array (1K) can cycle >1M times without any fails.
    • I can buy a 1Gbit chip today that can cycle >100K times (50nm SLC with ECC).
    • Planar TLC can cycle 3K times…. If I allow 1% failure, it last 10K cycles, the average (median) bit in that array lasts >30K cycles.
    • If I slow down the program and read timing dramatically, I can make a TLC 3K part last 10K+
    • And hypothetically, If I build a QLC SSD with massive redundancy and overprovisioning (think RAID), I can have 10 drive writes per day for 10 years. Looks like 36,500 cycles to end user.. but it is not

What does all this mean? The memory companies (and some SSD companies) know all of this and have details on tradeoff and how to manage them... Sorry, it's complex. Ask the experts how it works in a given application. Asking “how many cycles does that last?” or saying “QLC isn’t good enough” may not be useful

We can discuss specific questions to ask (BER, FIT, AFR, DPM, etc) and compare NAND, DRAM, MRAM, ReRAM, 3DXP in more detail. We published estimated 3DXP cycling performance numbers. Call for more information

next blog, I will answer the question "is Schrodinger's cat dead or alive??". As I tell my kids when we discuss at the dinner table .... "It's complex" 

Mark Webb





Friday, March 8, 2019

5 Things to Know About the Current Memory Market


5 Things to Know About the Current DRAM and NAND Market

Everyone is concerned about the memory market and trying to predict the future. Five Items to consider






1) Memory market is still growing long term. But the numbers matter. DRAM bits are growing at 18-20% CAGR. NAND bits are growing 35% CAGR. That is about the lowest CAGR in history, is indicative of a maturing market and includes effects of of all the buzzwords (SSD, NVMe, AI, ADAS, 5G, Edge Computing). What does this mean? On DRAM, we can’t always grow our way out of supply issues (without crashing the price) and if prices drop at all, it takes a long time for revenue to recover. On NAND, we can grow our way out but it requires massive price drops.

2) As mentioned in a previous note, elasticity helps growth but it is not as large as expected, not as quick as hoped, and whenever we are talking about significant elasticity, profits are at risk.

3) A famous man said recently “There is No Collusion!!”. IF the DRAM suppliers work together to never lower price and hold inventory until Apple, Dell, Google cave on pricing, then DRAM profits will continue at unreal levels. But even getting 3 people to collude appears to be tough… this is evident in DRAM pricing. The "MKW" report says “currently little evidence of collusion”. 

4) NAND is in a time where profitability is not a given. At this time, one of the keys is “who has lowest cost (full and cash), who can best survive a zero profit market?” Historically we would often say Samsung is the low cost producer. But differences in technology and the movement from 32L-128L has changed the leaders along the way. Also, target markets from phones, to Client SSDs, to Enterprise SSDs will change this.

5) When does this bottom out? Micron will probably give us the next checkpoint on Mar 20th. But we need to un-hide the demand by fixing inventory. If Amazon knows Hynix is holding 12 weeks of inventory, they will reduce theirs to “working inventory” and demand lower prices. Starting less wafers is not a smart idea so companies build more inventory. Once corrected, long term demand, short term demand from the field, plans for supply growth and revised cost numbers all kick in to tell us whether it is a 1 quarter dip or a 6 quarter dip. We have the numbers from all of these areas and an estimate on when it will recover that is updated weekly.

Mark Webb