Intel Shares New Roadmap for Optane, NAND, Including 144 Layer QLC and TLC

[Raevenlord]

Intel today at a press event in South Korea announced their plans for future product launches in the memory spaces. Optane is the name of the carriage Intel is pulling here - there's no novelty about that - and the company will be pushing a second generation release of Optane enterprise SSDs and Optane DC Persistent Memory modules. Most interesting for us down-to-earth PC enthusiasts, though - the market launch of 144 Layer QLC NAND in 2020, which should bring even lower pricing to NAND-based devices. Later, the company also plans to launch 144 layer TLC NAND solutions.

The new Optane modules apparently make use of first-generation 3D XPoint memory still - the love child of the now defunct Intel-Micron partnership. Intel's new Optane DC Persistent Memory products will materialize in codename Barlow Pass modules, with a release window around the likes of Cooper Lake (14nm) and Ice Lake (10nm) server processors scheduled for 2020. It seems that Intel's only consumer solution based in Optane - the Optane Memory H10 two-in-one SSD - is a lonely child effort which won't be joined by the previously-planned Optane Memory M15 (a dedicated cache drive for systems with mechanical-based storage, which are already on their way out) and Optane SSD 815P (which would only offer 118 GB of storage, clearly too little for current data storing trends in the overall market.





On to the SSD side of things, which what you're likely most interested in, Intel announced that they will be releasing a successor to their well-received 660p SSD late this year or early 2020, in the form of the 665p - which brings with it an even further reduction in pricing due to the updated 96-layer NAND (compared to the 64-layer currently employed in the 660p). Performance should remain around the same level - but price/bit is Intel's 660p strength after all.

Later in 2020, Intel will be launching updated SSD products which employ the company's 144 layer QLC NAND, bringing yet another nosedive in pricing. It's likely performance won't shift too much between their 96-layer SSD and their 14-layer one when it finally releases - perhaps Intell's betting on their 144-layer TLC to bring a new, higher-performing product to market, leaving the 665p's successor the task of picking up the bulk of storage space on users' computers due to its expected much lower pricing.

As a footnote, Intel also has demonstrated 5-bit-per-cell NAND flash memory working on its labs, but the company still doesn't know if the technology is feasible (cost and reliability-wise, it seems) to implement and bring to market

View at TechPowerUp Main Site

 

[trparky]

5-bit-per-cell NAND

Um... no thanks.

 

[dalekdukesboy]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

 

[Chaitanya]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

Intel has been making NAND for a very long time they used to make their own controllers as well but now they have stopped. Their NAND division has been doing much better than processor division. I think you must be living under a rock or bridge not knowing about IMFT/Intel NAND division.

 

[Crackong]

QLC? Nope

 

[dicktracy]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

Don't worry. Most people doesn't know what AMD is outside of their tiny circle jerk.

 

[bug]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

That's Intel's strength: you can beat it on one front, they will gather their resources from elsewhere and come back to bite your head off. It's why companies diversify.

 

[kapone32]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

Intel is into much more than just CPUs. They are actually one of the pioneers of NAND based hard drives as well. In terms of products outside of CPUs. I have 4 660P 1 TB NVME drives (which are the cheapest 5 yr warranty drives) and as much hate as they get they serve me fine and boot up every time. They are also huge in the WIFI world as their 8265 and 9260 adapters are some of the best PCI_E based cards you can get. Now I am also a proud owner of the AX200 which is a WIFI 6 adapter and even though I don't have a WIFI 6 router I see the benefits.

 

[dalekdukesboy]

That's Intel's strength: you can beat it on one front, they will gather their resources from elsewhere and come back to bite your head off. It's why companies diversify.

Agreed they are huge and I have an Intel Processor currently and have since core 2 duo, so I posted this as an Intel user and who have read the many reviews on their hard drives etc, so the Intel fanboi's circle jerking while climbing out from under their rocks to throw them at me from glass houses....is amusing . It's all good. You gave an intelligent response that made sense, hence me quoting only you however I will note no one in their hate actually addressed my point just attacked me personally in my opinion/knowlege, so...I struck a nerve and the Intel cult reacted.

 

[notb]

Intel is known for their processors, great, they have SSD's etc but...this smells like a desperate attempt to spin a positive bit of news while AMD is beating their ass into submission in CPU advances both on 7nm node vs 14nm and much greater architecture gains with Ryzen 3 etc. In other words, I don't think many if any in tech market give a flying fuck about optane/NAND etc at the end of the day.

You really shouldn't look at Intel with this kind of perspective.

AMD is a straightforward CPU and GPU maker. True.

Intel is not just a CPU company anymore. CPUs are their flagship product, but they're actually offering solutions with many types of products involved: CPUs, SSDs, network devices, mobile, Optane, accelerators, Thunderbolt3 and soon GPUs. Software as well: compilers/interpreters, libraries, ML, AI and so on.
I won't be even slightly shocked if Intel opens a consultancy firm in next 5 years. Or starts selling software.

As much as you may not be interested in Optane, it's very popular in enterprise sector, because it tries to solve one of the biggest problems: RAM capacity.
Intel will keep developing this idea as long as it's leading - hoping that it'll became so ubiquitous that future high-volume persistent "RAM" will either be based on Optane or at least fully compatible.

Nvidia has CUDA and Intel may have Optane.
It's not a technology that helps you win benchmarks when your flagship product is beating the competition.
It's a technology that will help your sales in the brief periods when the competition leaps ahead. Because you know they will from time to time.

 

[trparky]

Optane solves two of the biggest issues with NAND Flash memory.

1. Endurance
2. Low queue depth performance

Optane has far better endurance than NAND Flash meaning it can sustain thousands of more write operations than even MLC (Two-Bits-Per-Cell). It also performs faster than NAND Flash when faced with low queue depths like most users have. Sure, most NVMe SSDs boast high-performance numbers but those numbers only show in heavy I/O workloads. When it comes to real-world usage most SSDs are the same or within reason the same. This is an area that most SSD manufacturers haven't been able to improve performance in whereas Optane has shown major improvements.

 

[bug]

Optane solves two of the biggest issues with NAND Flash memory.

1. Endurance
2. Low queue depth performance

Optane has far better endurance than NAND Flash meaning it can sustain thousands of more write operations than even MLC (Two-Bits-Per-Cell). It also performs faster than NAND Flash when faced with low queue depths like most users have. Sure, most NVMe SSDs boast high-performance numbers but those numbers only show in heavy I/O workloads. When it comes to real-world usage most SSDs are the same or within reason the same. This is an area that most SSD manufacturers haven't been able to improve performance in whereas Optane has shown major improvements.

This. A thousand times this. If only Intel could get the power draw (and cost) under control...

 

[trparky]

Yep. If you ever read some benchmarks there's talk about something called "4K random reads". The thing about "4K random reads" is that this is the benchmark that most closely mirrors that of real-world performance. Most read operations are going to be quite small, usually in blocks of 4 KBs and they usually are very random hence the idea that "4K random reads" most closely mirrors that of real-world numbers.

For the longest time, most SSDs share one thing in common... They all share similar "4K random reads" benchmark numbers. Okay sure, sustained numbers may indicate better results but they don't indicate what real performance is going to be like for 99% of users unless of course they're writing and reading huge 4K video files.

 

[bug]

Yep. If you ever read some benchmarks there's talk about something called "4K random reads". The thing about "4K random reads" is that this is the benchmark that most closely mirrors that of real-world performance. Most read operations are going to be quite small, usually in blocks of 4 KBs and they usually are very random hence the idea that "4K random reads" most closely mirrors that of real-world numbers.

For the longest time, most SSDs share one thing in common... They all share similar "4K random reads" benchmark numbers. Okay sure, sustained numbers may indicate better results but they don't indicate what real performance is going to be like for 99% of users unless of course they're writing and reading huge 4K video files.

Which wouldn't be such a problem if we didn't have sequential numbers of ~3GB/s while in the real world we're limited by 4k random reads which are well below 100MB/s. I believe the best AHCI drives can do ~50MB/s, whereas NVMe can do ~70MB/s.

Sequential speeds can also come in handy when doing backups and restores. But you'd have to connect your backup storage using something equally fast... If not, video and photo editing could benefit from that.

 

[trparky]

Hence Optane...

 

[bug]

Hence Optane...

It would improve on SSD's weak points, but Optane hasn't gone anywhere in the past couple of years. To be fair, CD writers were also >$1,000 a piece for over a decade before becoming mainstream (and eventually going the way of the dodo) so who knows?