Slashdot Mirror
DRAM Almost as Fast as SRAM
An anonymous reader writes "IBM said it has been able to speed up the DRAM to the point where it's nearly as fast as SRAM, and that the result is a type of memory known as embedded DRAM, or eDRAM, that helps boost the performance of chips with multiple core calculating engines and is particularly suited for enabling the movement of graphics in gaming and other multimedia applications. DRAM will also continue to be used off the chip."
to go for title of most patents filed in 2007
"Stallman says add to this code and you are one of us. Gates says use this code and you belong to us."
Firstly, system memory is not especially fast compared to the CPU, and the recent proliferation of multiple cores is making the situation worse because more CPUs are trying to bang on the same memory.
Secondly, the most straightforward way to paper over problems with high-latency devices is to put a cache in front of them. Super fast DRAM would be one way to enable bigger caches that reduce the impact of various system bottlenecks. Sure we can hope to replace all hard drives with solid state devices, but since they still cost orders of magnitude more per megabyte, it will probably be quite a while before that happens. In the mean time, better caches couldn't hurt.
To those wondering why it would be good to have DRAM as fast as SRAM: SRAM doesn't need to be "refreshed" constantly, and is faster, but takes up many more transistors and is therefore much less dense and more expensive for the same amount of memory.
However with DRAM it takes quite a bit of power just to keep data in memory (because of the constant "refreshes"), which isn't the case with SRAM. So this discovery wouldn't take SRAM out of production for applications which require its low power usage.
// MD_Update(&m,buf,j);
Because your spinning magnetic platter is a cheaper storage "solution" than edram, flash, whatever.
Unless you want to pay $25 per GB [again...], I'd wait until things improve.
And it isn't like they're not working on smaller/faster memory. Two years ago a 1GB flash was 99$ [in Canada], now they're ~40$ and you can get a 2GB flash for about the price of the 1GB. I imagine this year we'll see 4GB flash drives become more of a norm, and so on.
Most likely, ten years from now 80GB flash drives will be common place enough and not super expensive. But until then, spinning platers!
Someday, I'll have a real sig.
Why not concentrate on the bottlenecks
In comparison to the processor, is RAM not a bottleneck? An improvement in an area that has less need is still an improvement.
I don't get why this is news. Embedded-DRAM has been in heavy usage for many years now.
Both the title and the summary are quite misleading, since eDRAM is on-chip and that of course is much faster than external off-chip memory, be SRAM, DRAM or whatever.
Some big examples? PS2, Nintendo Gamecube, Wii, Xbox 360. All these consoles use eDRAM for their GPU's on-chip framebuffers to enhance their performance, and that goes back to at least the year 2000 when the PS2 came out.
Some will be quick to say "no, the Nintendo consoles use 1T-SRAM, not DRAM". Yeah, right, but even 1T-SRAM (despite its name) is a form of embedded-DRAM.
- Otaku no naka no otaku, otaking da!!!
EE Times article. Today SRAM is used for processor caches, but new multicore chips need massive (i.e. expensive) cache. Because eDRAM is much denser than SRAM, it allows chip designers to fit much more cache in the same size chip, increasing overall performance. IBM and AMD use silicon-on-insulator (SOI) technology, while the rest of the industry uses bulk CMOS; eDRAM for bulk has been available for a while (it's used in Xbox 360 and BlueGene/L for example), but now IBM has developed SOI eDRAM that can be used in IBM's future processors (and maybe AMD's).
Prices too high, sizes too small:
2 E168201631592 E168202201560 03240522+1309421175&Submit=ENE&SubCategory=522
http://www.newegg.com/Product/Product.asp?Item=N8
http://www.newegg.com/Product/Product.asp?Item=N8
http://www.newegg.com/Product/ProductList.asp?N=2
4GB flash for $40-$60, sd for $45, so $10-$15 per GB, right now. 1 GB cost $60 about 18 months ago(they are less than $15 now); extrapolate linearly, thats 64GB for cheap($60!) in 6 years, and 128+ in 8 years. That doesn't account for a slight depression in prices as the size of the chips used goes up.
I'd pay $100 extra for a laptop with a 32GB flash drive to go with the giant hard disk, just to save power. That's fairly likely in less than 4 years.
Nerd rage is the funniest rage.
There are 2 areas of latency for a cache, the first is the performance of the actual data cells, and the second is the speed of doing a lookup in the cache. The larger the cache, and the higher the degree of set associativity, the longer the lookup takes. Thus you're unlikely to see this eDRAM used for L1 caches, and probably not for L2 caches either, as more cache would slow them down, even if the cells are just as fast as SRAM. The sweet spot will probably be for L3 caches, that are already slow by cache standards, but a whole lot faster than system memory. Since L3 caches are large, the cost savings for switching to eDRAM would be largest there.
As for power concerns, DRAM is higher than SRAM, but a larger L3 cache may reduce the traffic through the memory controller, and out to the DIMMs, which will probably more than make up for any increase in power density in the cache.
No, I'm not a fan of patent trolls; but this isn't patent trolling. IBM has created a new, better way to embed cache RAM on the CPU die, at a signifigant cost in both manpower and materiel. This isn't like they patented "a method to check customers out with one click" or something similarly banal. This is a real, new technology which took a great deal of time, energy and work to create. No "prior art", no "trivially obvious" - this is exactly the kind of technological advancement which patents should protect.
I expect to see 80GB flash drives long before 10 years. Assuming a growth rate of doubled capacity every 18 months, true enough, we'd reach about 80 GB in 10 years, but so far, flash memory has increased much faster than Moores law. Also I assume that the amount of data our computers manipulate continue to increase with each version of windows/HD-DVD/whatever, so we still need larger/slower storage mediums in 10 years, such as harddisks.
In fact, the whole idea of using a (set of) rotating platter(s) with magnetic coating and radially movable read/write head(s) for storage, has been so successful for so long, and continue to improve at such an astonishing rate, that I doubt it will go away any time soon. In the far future, it's more difficult to predict what would happen. But even today, wheels are important, fire is our main source of (non-food) energy, primitive cutting tools are regulary used in any household, and in general, assuming things fail to change, is rarely wrong (we still haven't got flying cars!)
Yeah, but wouldn't it be better to buy a real computer with room for more RAM, so you didn't have to use a hardware device to imitate another hardware device, so that you could use software to imitate the drivers of the other hardware device, so that you could use it as the first kind of hardware device, just with lower speed and convenience? Or in other words: wouldn't it be better to just run the database in RAM?
I am in no way an expert, but I read about other upcoming types of RAM which also sound interesting:
Z-RAM. One cell is a single transistor. Faster than SRAM, which uses 6 transistors per cell. http://en.wikipedia.org/wiki/ZRAM
TTRAM. One cell contains 2 transistors. As fast as SRAM, according to Wikipedia. http://en.wikipedia.org/wiki/TTRAM