Tuesday, July 28th 2015

Intel and Micron Produce Breakthrough Memory Technology

Intel Corporation and Micron Technology, Inc. today unveiled 3D XPoint technology, a non-volatile memory that has the potential to revolutionize any device, application or service that benefits from fast access to large sets of data. Now in production, 3D XPoint technology is a major breakthrough in memory process technology and the first new memory category since the introduction of NAND flash in 1989.

The explosion of connected devices and digital services is generating massive amounts of new data. To make this data useful, it must be stored and analyzed very quickly, creating challenges for service providers and system builders who must balance cost, power and performance trade-offs when they design memory and storage solutions. 3D XPoint technology combines the performance, density, power, non-volatility and cost advantages of all available memory technologies on the market today. The technology is up to 1,000 times faster and has up to 1,000 times greater endurance3 than NAND, and is 10 times denser than conventional memory.
"For decades, the industry has searched for ways to reduce the lag time between the processor and data to allow much faster analysis," said Rob Crooke, senior vice president and general manager of Intel's Non-Volatile Memory Solutions Group. "This new class of non-volatile memory achieves this goal and brings game-changing performance to memory and storage solutions."

"One of the most significant hurdles in modern computing is the time it takes the processor to reach data on long-term storage," said Mark Adams, president of Micron. "This new class of non-volatile memory is a revolutionary technology that allows for quick access to enormous data sets and enables entirely new applications."

As the digital world quickly grows - from 4.4 zettabytes of digital data created in 2013 to an expected 44 zettabytes by 2020 - 3D XPoint technology can turn this immense amount of data into valuable information in nanoseconds. For example, retailers may use 3D XPoint technology to more quickly identify fraud detection patterns in financial transactions; healthcare researchers could process and analyze larger data sets in real time, accelerating complex tasks such as genetic analysis and disease tracking.

The performance benefits of 3D XPoint technology could also enhance the PC experience, allowing consumers to enjoy faster interactive social media and collaboration as well as more immersive gaming experiences. The non-volatile nature of the technology also makes it a great choice for a variety of low-latency storage applications since data is not erased when the device is powered off.

New Recipe, Architecture for Breakthrough Memory Technology
Following more than a decade of research and development, 3D XPoint technology was built from the ground up to address the need for non-volatile, high-performance, high-endurance and high-capacity storage and memory at an affordable cost. It ushers in a new class of non-volatile memory that significantly reduces latencies, allowing much more data to be stored close to the processor and accessed at speeds previously impossible for non-volatile storage.

The innovative, transistor-less cross point architecture creates a three-dimensional checkerboard where memory cells sit at the intersection of word lines and bit lines, allowing the cells to be addressed individually. As a result, data can be written and read in small sizes, leading to faster and more efficient read/write processes.

3D XPoint technology will sample later this year with select customers, and Intel and Micron are developing individual products based on the technology.


Add your own comment

19 Comments on Intel and Micron Produce Breakthrough Memory Technology

#1
emissary42
As the digital world quickly grows - from 4.4 zettabytes of digital data created in 2013 to an expected 44 zettabytes by 20204
Then i guess, there still is a lots of time for new technologies like this^^

There also is a 3 right after endurance, so maybe these are footnote-remains?
Posted on Reply
#2
DeathtoGnomes
is this implying bigger better faster flash drives?
Posted on Reply
#3
64K
Up to 1,000 times faster. Up to 1,000 times greater endurance. 10 times denser.
Well, damn! That is indeed a breakthrough.

A decade of research. I wonder what is just a gleam in the eye of some engineer/scientist right now that will pop up a decade from now after research that will be truly amazing.
Posted on Reply
#4
profoundWHALE
I'm not worried about Intel/Micron delivering the product in terms of it's capabilities. I am worried about them producing it in high enough quantities, similar to the problem that AMD faced.
Posted on Reply
#5
RejZoR
I hope it doesn't have retarded limited number of write cycles and idiotic need for TRIM. NAND sucks balls because of these two...
Posted on Reply
#6
Petey Plane
profoundWHALE said:
I'm not worried about Intel/Micron delivering the product in terms of it's capabilities. I am worried about them producing it in high enough quantities, similar to the problem that AMD faced.
As the digital world quickly grows - from 4.4 zettabytes of digital data created in 2013 to an expected 44 zettabytes by 20204
Don't worry, they have 18189 years to perfect the manufacturing process

edit: aw, emissary42 beat me to the joke
Posted on Reply
#7
Flibolito
If true Ram will end up being your local storage. This could be a epic transition. Imagine a graphics card 5-7 years from now where you just install the game onto it and let it run directly on the card. Time will tell, might be time to buy some stock.
Posted on Reply
#8
RejZoR
Unified operational and storage memory is the future. It's the only way to achieve fast data access and processing. Keeping them separate will ALWAYS cause bottlenecks.
Posted on Reply
#9
profoundWHALE
RejZoR said:
Unified operational and storage memory is the future. It's the only way to achieve fast data access and processing. Keeping them separate will ALWAYS cause bottlenecks.
Source: https://en.wikipedia.org/wiki/Von_Neumann_architecture#Von_Neumann_bottleneck
Because program memory and data memory cannot be accessed at the same time, throughput is much smaller than the rate at which the CPU can work. This seriously limits the effective processing speed when the CPU is required to perform minimal processing on large amounts of data.
Posted on Reply
#10
sfp1987
This sounds like a Memristor technology.
Posted on Reply
#12
R-T-B
Flibolito said:
If true Ram will end up being your local storage. This could be a epic transition. Imagine a graphics card 5-7 years from now where you just install the game onto it and let it run directly on the card. Time will tell, might be time to buy some stock.
It's nonvolatile. You lose power, you reinstall...

EDIT: wait... It's late I'm sorry. Nonvolatile means the opposite of that, lol...
Posted on Reply
#13
Jatheon
DeathtoGnomes said:
is this implying bigger better faster flash drives?
Good question! We also expect higher speed.
Posted on Reply
#14
Aquinus
Resident Wat-man
btarunr said:
The technology is up to 1,000 times faster and has up to 1,000 times greater endurance3 than NAND, and is 10 times denser than conventional memory.
Typo or are we missing some fine print?
profoundWHALE said:
Source: https://en.wikipedia.org/wiki/Von_Neumann_architecture#Von_Neumann_bottleneck
With memory alone at any given instance, sure. Several levels of cache mitigate this problem and if it really were a bottleneck with current technology, faster memory would be yielding tangible performance benefits, which it doesn't. Also shared memory is always required at some level as in the end there always needs to be some level of shared state between cores in a multi-threaded/process environment.
Posted on Reply
#15
lilhasselhoffer
So, the part of me that wants to be contrarian is calling bullshit at the start. A quick check of the forum segment...yep, it's a press release. Prepare your grains of salt now.


It'll take a bit of effort to justify myself, but it seems like Intel has already done the work for me... Sigh...
http://www.bbc.com/news/technology-33675734
http://www.micron.com/about/innovations/3d-xpoint-technology


The highlights of the two articles (one from the BBC, the other from Micron) are as follows:
1) This won't be replacing NAND any time soon, primarily because of cost.
2) The chips they've got are faster than NAND, but slower than RAM.
3) The chips apparently cost less than the manufacture of RAM.
4) The primary current benefit to this technology is in sheer volumes. Think about a games server with no RAM, but two or three times the amount of this memory acting as a RAM substitute. Those 128 GB servers are now up to 256 GB or more memory, which amounts to more players, on way less servers, with lower power draws.




The 1000's of times is when the technology is significantly more matured, and not what we'd be seeing in the next few years. It's functionally extrapolating the future from potential mathematical models. In Intel's discussion with the BBC they're functionally suggesting it's to be used as a slower but bigger cache. Think about it like the hybrid SSD/HDD drives of the past. You're not as fast as the fastest stuff, but you get better bang for the monetary investment.
Posted on Reply
#16
Frick
Fishfaced Nincompoop
So how does it work?
Posted on Reply
#17
lilhasselhoffer
Frick said:
So how does it work?
That BBC article I linked to has a good explanation.

In short terms, it's a pair of perpendicularly arranged wire meshes, with a column between them. Each colum has a selector, and a storage segment. The storage segment is a memresistor. It's funny, but the BBC states exactly that while somehow forgetting such a thing was huge news only a couple of years ago. By passing through a given amount of power the storage segment varies in resistance, representing the 0 and 1. The intersection of each perpendicular wire only occurs once, so a single bit can be written at a time, and because the wires are perpendicular they can theoretically be stacked into a huge conglomeration (think something like the finfet transistors Intel previously introduced, only actually used in 3d).
Posted on Reply
#18
Frick
Fishfaced Nincompoop
@lilhasselhoffer Thanks! Somehow it's nice they didn't drum up the memristor part, it feels more real that way. That explains the high costs too, I don't think anyone is making them en masse yet.
Posted on Reply
#19
Prima.Vera
Sorry, it says the "Video is private".

Cheers.
Posted on Reply
Add your own comment