Toshiba's 768Gb 3D QLC NAND Flash Memory: Matching TLC at 1000 P/E Cycles?
by Anton Shilov on July 3, 2017 2:30 PM ESTToshiba last week announced its first 3D NAND flash memory chips featuring QLC (quadruple level cell) BiCS architecture. The new components feature 64 layers and developers of SSDs and SSD controller have already received samples of the devices, which Toshiba plans to use for various types of storage solutions.
Toshiba’s first 3D QLC NAND chips feature 768 Gb (96 GB) capacity and uses 64 layers, just like the company’s BICS3 chips with 256 Gb and 512 Gb capacities launched in 2016 and 2017. Toshiba does not share further details about its 3D QLC NAND IC (integrated circuit), such as page size, the number of planes as well as interface data transfer rate, but expect the latter to be high enough to build competitive SSDs in late 2018 to early 2019 (that’s our assumption). Speaking of applications that Toshiba expects to use its 3D QLC NAND ICs, the maker of flash memory mentions enterprise and consumer SSDs, tablets and memory cards.
Endurance++
Besides intention to produce 768 Gb 3D QLC NAND flash for the aforementioned devices, the most interesting part of Toshiba’s announcement is endurance specification for the upcoming components. According to the company, its 3D QLC NAND is targeted for ~1000 program/erase cycles, which is close to TLC NAND flash. This is considerably higher than the amount of P/E cycles (100 – 150) expected for QLC by the industry over the years. At first thought, it comes across a typo - didn't they mean 100?. But the email we received was quite clear:
- What’s the number of P/E cycles supported by Toshiba’s QLC NAND?
- QLC P/E is targeted for 1K cycles.
It is unclear how Toshiba managed to increase the endurance of its 3D QLC NAND by an order of magnitude versus initially predicted. What we do know is that signal processing is more challenging with QLC than it is with TLC, as each cell needs to accurately determine sixteen different voltage profiles (up from 2 in SLC, 4 in MLC, and 8 in TLC).
The easiest way to handle this would be to increase the cell size: by having more electrons per logic level, it is easier to maintain the data and also read from it / write to it. However, the industry is also in a density race, where bits per mm^2 is an issue. Also, to deal with read errors from QLC memory, controllers with very advanced ECC capabilities have to be used for QLC-based SSDs. Toshiba has its own QSBC (Quadruple Swing-By Codes) error correction technique, which it claims to be superior to LDPC (low-density parity-check) that is widely used today for TLC-powered drives. However, there are many LDPC implementations and it is unknown which of them Toshiba used for comparison against its QSBC. Moreover, there are more ECC methods that are often discussed at various industrial events (such as FMS), so Toshiba could be using any or none of them. The only thing that the company tells about its ECC now is that it is stronger than 120 bits/1 KB used today for TLC. In any case, if Toshiba’s statement about 1000 P/E cycles for QLC is correct, it means that that the company knows how to solve both endurance and signal processing challenges.
The main advantage of QLC NAND is increased storage density when compared to TLC and MLC, assuming the same die size. As was perhaps expected, die size numbers were not provided. However, last year Toshiba and Facebook talked about a case study QLC-powered SSD with 100 TB of capacity for WORM (write once read many) applications and it looks like large-capacity custom drives and memory cards will be the first to use QLC for cold storage. P/E cycles and re-write endurance isn't a concern for WORM at this stage.
Toshiba has begun to sample its 3D QLC NAND memory devices earlier this month to various parties to enable development of SSDs and SSD controllers. Taking into account development and qualification time, Toshiba plans to mass produce its BiCS3 768 Gb 3D QLC NAND chips around the same time it starts to make its the next generation BiCS4 ICs. The latter is set to hit mass production in 2018, but the exact timeframe is yet to be determined.
Related Reading:
- Western Digital Announce BiCS4 3D NAND: 96 Layers, TLC & QLC, Up to 1 Tb per Chip
- Toshiba Selects Japan-U.S. Consortium As Preferred Buyer For Memory Business
- Toshiba Introduces XG5 Client NVMe SSD
- Western Digital Ships Client SSDs Based on 512 Gb 3D TLC NAND Chips
- Toshiba Samples 64-Layer 512 Gb BiCS 3D NAND, Announces 1 TB BGA SSD
- Toshiba Finalizes Plans for New 3D NAND Fab: Coming Online in 2019
Source: Toshiba
36 Comments
View All Comments
Alexvrb - Wednesday, July 5, 2017 - link
What firm do you do engineering work for again? May I see your whitepapers? With the move to 3D NAND I have no issues with good TLC drive. QLC... fine for storage, at the very least. At 1000 P/E it will likely have a lower average failure rate than HDDs, even for primary storage - not that I would use it for primary storage.BurntMyBacon - Monday, July 17, 2017 - link
@AlexvrbWhat part of "Unfortunately, despite all the data that suggests TLC has plenty of endurance for consumers, I can't bring myself to put a TLC drive in my personal system." suggests that this is anything more than an opinion.
@Alexvrb: "What firm do you do engineering work for again? May I see your whitepapers?"
What exactly do you want me to prove here. I've already stated that there is plenty of data that suggests TLC had plenty of endurance for consumers. I didn't mention enterprise, but if you are talking enterprise, there are some use cases that require eMLC. If regular MLC isn't up to the task, then clearly TLC won't be for that scenario. If you are talking about the error correction method, there are plenty word dedicated to that in the article above.
From Article
"Also, to deal with read errors from QLC memory, controllers with very advanced ECC capabilities have to be used for QLC-based SSDs. Toshiba has its own QSBC (Quadruple Swing-By Codes) error correction technique, which it claims to be superior to LDPC (low-density parity-check) that is widely used today for TLC-powered drives."
Are you suggesting that I need to show you white paper before I can have an opinion of what performance and endurance I am willing to pay for? I tell my customers that there is minimal risk with a good TLC drive and a little more so with cheaper drives like the Kingston UV400, but controller failure often happens before the NAND wears out anyways. Furthermore, for most people, even the cheapest TLC drive last much longer than spinning media. If a client wants a QLC drive, I'll give it to them. I will not, however, suggest it to the client as at this point there are too many unknowns. A NAND endurance of 1000 P/E cycles is more than some TLC drives (Kingston UV400 gets half that). That said, it has not been tested and it is unknown whether they can actually pull it off. Also consider, that the number relies on advanced ECC capabilities. The effect this has on the controller is an unknown quantity. There were drives in TechReport's SSD Endurance test that failed without a single bad sector, suggesting the controllers were responsible:
http://techreport.com/review/27909/the-ssd-enduran...
Full Disclosure: I have designed ASICs fabricated for and in use at Washington University's Radio Chemistry department (nuclear physics), so while I have no experience design pure memory chips, I do know a few things about silicon charge decay rates. Do you have any credentials to provide or did you think you could give yourself credibility by discrediting me?
grant3 - Friday, July 7, 2017 - link
Consumer SSDs are regularly being endurance tested, and invariably are proven to handle more wear than even a power user would put on a drive before it's hopelessly obsolete. There is no "trade-off" for someone buying a TLC drive unless they're sticking it in a datacenter.The odds of the drive being lost to fire, theft, or alien abduction are much greater than some some crazy random chain of cell failures overpowering normal error correction.
A responsible advisor would simply tell his clients to regularly back up their data on PROPER archive media, and use the money they saved on some beers so they can relax.
BurntMyBacon - Monday, July 17, 2017 - link
@grant3: "Consumer SSDs are regularly being endurance tested, and invariably are proven to handle more wear than even a power user would put on a drive before it's hopelessly obsolete."Agreed. Quick question: Do you have a line on endurance tests with anything newer than a Samsung 840 EVO? My 60 second GoogleFu has failed me in this regard. Of particular interest to me is offline data retention, which happens to be the number 1 storage failure issue I personally encounter with SSDs.
@grant3: "The odds of the drive being lost to fire, theft, or alien abduction are much greater than some some crazy random chain of cell failures overpowering normal error correction."
While the odds are great, I submit for you a counter example:
http://techreport.com/review/27909/the-ssd-enduran...
"The 840 Series didn't encounter actual problems until 300TB, when it failed a hash check during the setup for an unpowered data retention test. The drive went on to pass that test and continue writing, but it recorded a rash of uncorrectable errors around the same time. Uncorrectable errors can compromise data integrity and system stability, so we recommend taking drives out of service the moment they appear."
The workload is not a client workload, so this really isn't something to worry about. However, it proves that it can happen. By virtue of requiring stronger error correction, it stands to reason that QLC based drives are more likely to see this type of unrecoverable error than TLC drives. The process size the NAND is built on has also shrunk since then. Shrink the process a few more times and the odds may not be as far fetched as they once were.
@grant3: "A responsible advisor would simply tell his clients to regularly back up their data on PROPER archive media, and use the money they saved on some beers so they can relax."
Almost exactly what I tell them word for word. Almost never happens outside of businesses. They tell me up front they won't do it. I'll still put a TLC drive in as they usually last longer than spinning disk. However, given this situation QLC doesn't inspire as much confidence as I'd like (this is an opinion). One client I had with a proper backup solution in place still lost critical data because he relaxed the period between backups due to perceived performance issues on the network that were completely unrelated.
vaibhav24 - Wednesday, April 10, 2019 - link
3D NAND Flash Memory Market Analysis - http://bit.ly/2KyjJXY3D NAND Flash Memory Report by Material, Application, and Geography – Global Forecast to 2022 is a professional and in-depth research report on the world's major regional market conditions, focusing on the main regions (North America, Europe and Asia-Pacific) and the main countries (United States, Germany, united Kingdom, Japan, South Korea and China).
The report firstly introduced the 3D NAND Flash Memory basics: definitions, classifications, applications and market overview; product specifications; manufacturing processes; cost structures, raw materials and so on. Then it analyzed the world's main region market conditions, including the product price, profit, capacity, production, supply, demand and market growth rate and forecast etc. In the end, the report introduced new project SWOT analysis, investment feasibility analysis, and investment return analysis.