Slashdot Log In
Intel Set To Demo PRAM
Posted by
kdawson
on Tue Apr 17, 2007 05:17 AM
from the flashier dept.
from the flashier dept.
xavatarx writes "Intel's chief technology officer Justin Rattner is set to give the first public demonstration of the company's PRAM (phase-change RAM) technology at this week's Intel Developer Forum conference. 'Intel and other companies are counting on PRAM to replace both NOR and NAND flash memory to generate the demand required to produce the new memory chips in volume, and drive down costs,' the article says."
Related Stories
[+]
Intel to Sample Flash-killer PRAM This Year 78 comments
Station writes "Intel's new phase-change memory technology (PRAM) will begin sampling this year. Samsung, IBM, and Hitachi are all working on phase-change memory as a successor to flash as it has a lower (~20ns) read latency than flash (50-90ns). 'Intel says they plan to ship the first PRAM modules as a straight-ahead NOR flash replacement so that they can work the kinks out of the design before trying to move it up the memory hierarchy. The company claims a much higher number of read-write cycles (100 million) than flash, as well as a potential 10 years' worth of data retention. NOR flash is typically used as program storage memory for mobile devices like cell phones, while more durable but slower NAND flash is used for mass storage in devices like the iPod nano.'"
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Quite the demo (Score:5, Funny)
Hope it is better than Intel's other memory push (Score:3, Insightful)
AMD set to reply (Score:3, Funny)
PRAM is new? (Score:5, Funny)
Re:PRAM is new? (Score:5, Informative)
Apple PRAM == Parameter RAM
Intel PRAM == Phase-change RAM
While:
Parameter RAM == Any kind of conventional (probably non volatile) RAM
Phase-change RAM == New kind of non volatile RAM using a new phase change technology
Thus:
Apple PRAM != Phase-change RAM
QUED.
Parent
Question on how PRAM works and is manufactured (Score:5, Interesting)
How are Intel and others managing this chalcogenide glass manufacturing in their usual silicon DRAM process? Is this glass fused/bonded to silicon or something? Or is it an alloy.. and if so, is it a non-silicon alloy (silicon is a non-metal)?
I tried the wikipedia entry on this subject (http://en.wikipedia.org/wiki/Phase-change_memory
a. How it really works in an electronic circuit and in a microprocessor (how do you control the heating/cooling at the chip level so that phase change occurs)?
b. How it is supposed to be volume manufactured? Would they require a new fab entirely to manufacture PRAM (if they do decide to commercialize this technology), or can an existing fab be retro-fitted to support this manufacturing process?
Appreciate any insights on this subject. At a high level, this does sound like a very exciting new technology.
Parent
Re:Question on how PRAM works and is manufactured (Score:5, Informative)
http://www.ovonyx.com/tech_html.html [ovonyx.com]
It sounds like the chalcogenide is deposited as a thin film. Mind you they talk about transistors, so it must be an extra processing stage on a normal chip.
This seems to confirm it
http://www.eetimes.com/in_focus/silicon_engineeri
Chalcogenide RAM is nonvolatile, boasts access speed comparable to that of DRAM and possesses advantages in scalability, high sensing margin, low energy consumption and endurance to cycling. The structure and processing of chalcogenide memory are much simpler than in other next-generation memories such as MRAM and ferroelectric RAM. In a chalcogenide memory cell, the data is stored in a flat chalcogenide layer that can be deposited near the end of the CMOS interconnect process. Therefore, disturbance of the CMOS process is minimal, making it ideal for systems-on-chip.
So I guess they add an extra step to the end of the process and deposit a layer of chalcogenide glass.
These things sound really cool BTW, they're writable at a byte granularity in tens of nanoseconds just like a regular SDRAM, but they are non volatile. It looks like they can flip bits individually either way too.
Whereas flash memory is much slower - tens of microseconds per byte, and you need to erase 16K-128Kbyte block at a time. And PRAM is supposed to be denser and allow unlimited erase cycles.
Plus Intel is backing it so it's not like it will fail because the vendor can't afford to scale the production process to make chips with a high capacity.
Parent
Re:Question on how PRAM works and is manufactured (Score:5, Informative)
Parent
Re:Question on how to pronounce that word (Score:3)
chal-ko-gen'-ide ?
chal-cog'-en-ide ?
sh-alko'-gen-ide ?
fawstuh-grhants-dhaling ?
Re:Question on how to pronounce that word (Score:4, Funny)
Chris Mattern
Parent
Re:Question on how PRAM works and is manufactured (Score:5, Funny)
Parent
Re:Question on how PRAM works and is manufactured (Score:5, Funny)
What!? Don't tell me you're scared of fringe effects on leakage current, non-linear C-V models, and photo mask correction factors! Sheesh! Nerds these days...
Parent
Re:PRAM is new? (Score:4, Funny)
I suggest you add him to a list of humans that use humour, and ensure your emotion emulator is powered up before replying to posts by humans on that list in future. It is very important that the humans here do not realise that we are not like them.
Parent
Re: (Score:3, Informative)
Re:PRAM is new? (Score:5, Funny)
Parent
Re:PRAM is new? (Score:5, Funny)
Parent
Again? (Score:5, Interesting)
BTW, is there anywhere a concise table comparing the characteristics of different RAM technologies?
Re:Again? (Score:5, Informative)
The only thing not mentioned is cost, but if it scales (Intel thinks it does), and the special glass just adds only one step to manufacturing (which the original developer claims), it doesn't look like there are too many roadblocks to success. Now, claiming it will replace DRAM may be a little premature, however there's a good hope they can replace flash at least.
Parent
Re:Again? (Score:4, Informative)
Parent
How? Intel is a supermassive chip company w/FABS. (Score:5, Insightful)
There are 2 ways they can win assuming they have the capacity for massive scale manufacturing, which they seem to.
1 - They can undercut the cost of NOR/NAND chips in the market place.
2 - they can sell at price parity with NOR/NAND but solve the serious technical problems with both designs. TBH neither is terribly nice to interface to and both are very slow compared to DRAM. Beating the technical problems are explained to PHB's as Instant Boot (suspend to RAM - except its PRAM, not battery backed DRAM). Quietness. The G shocks required to actually damage the stuff as compared to a spinning HD. etc etc...
Actually there's a lot of reasons to go for it. Not least of all that HD sizes on consumer products (cheapo laptops etc) seem to be stuck at 80 gig. It's cheap and most (not all - I've a terabyte+ of storage and still run out) users will never fill it anyway. Instant boot is a real serious seller - 2 minutes of boot time feels a lot longer as you can't do anything for 2 minutes.
But, personally, I think Intel's massive production capability and their endless search to find something to do with all those billions of transistors is where they'll "win".
Parent
Well, why not. (Score:2)
Oh, wait...
Clarification (Score:5, Informative)
I'm generally more a software person than hardware, but there's a lot to be excited about with this. It's apparently got a r/w time only 2-3 times the time of DRAM, and holds a lot of potential for things like paging files and storing frequently used software since there doesn't seem to be a limitation on the number of writes that can be applied. Once things develop, the technology might even be a ludicrous-speed replacement for hard-drives, as the storage mechanism is quite stable (more so than flash). I can definitely see this taking off in the future, if it delivers what it promises (and nothing else supercedes it in the meantime).
Here's the issue (Score:5, Insightful)
A interesting example is the IBM 2301 drum memory device. Originally used as main memory, 2301's were later converted to paging devices. They had great transfer rates, but they became obsolete as soon as RAM sizes increased enough to cache a reasonable number of pages.
The reason is that even though "slow RAM" like drum memory seems intuitively useful as a "third stage" paging device, if you do the math versus the two-stage combination of very fast RAM and very much slower disk, you find that the RAM/disk combo performs almost as well. The conclusion therefore has been that it makes little sense to throw away your money on medium-speed RAM, because you'd have gotten more bang for the buck by spending it either on (a) more fast main memory, or (b) bigger/better disks.
Finally, if you look at history, the rotating storage industry continues to confound all of these "fast RAM" technologies by increasing performance and dropping $/bit at an amazing rate. Nothing is more primitive, to my mind, than spinning a disk platter in 2007 -- but there's still nothing better, and the technology shows no sign of dying.
Parent
Re: (Score:3, Insightful)
In working with large numbers of computers over the years, here are my informal statistics for failure. 1) hard drives 2) power supplies 3) other, almost at the noise level
And look at what the industry adds redundancy for in computers. 1) hard drives and 2) power supplies.
Sure, anyone that thinks about it will agree that spinning disk platters in 2007 i
Re: (Score:3, Interesting)
I would love to see the hard drive industry start incorporating more read heads in to drives.
Imagine a single drive with two armatures, servo's and read/write heads. You could effectively double the throughput, reduce seek times, and improve latency. It wou