Breaking News
News & Analysis

Memory-cube group issues initial draft specs

8/14/2012 12:00 PM EDT
8 comments
NO RATINGS
More Related Links
View Comments: Threaded | Newest First | Oldest First
resistion
User Rank
Author
re: Memory-cube group issues initial draft specs
resistion   8/14/2012 12:34:37 PM
NO RATINGS
Seems the TSV bandwidth is wasted on flash.

zanfar
User Rank
Author
re: Memory-cube group issues initial draft specs
zanfar   8/14/2012 3:20:11 PM
NO RATINGS
It was my understanding that the HMC consortium was primarily concerned with DRAM memory--not Flash. http://www.hybridmemorycube.org/technology.html

rick merritt
User Rank
Author
re: Memory-cube group issues initial draft specs
rick merritt   8/14/2012 6:29:00 PM
NO RATINGS
Yes the Memory Cube is for DRAM not flash. My mistake. The story above has been corrected.

Rchandta1
User Rank
Author
re: Memory-cube group issues initial draft specs
Rchandta1   8/14/2012 9:23:45 PM
NO RATINGS
Agreed stacking DRAMs has advantage. But I can't understand the rest. It is tightly coupled with the memory controller, so there is power and latency benefits. But it seems to force the processor away from the memory controller. Many processors have integrated DRAM controllers; so the stacking scheme seems to negate their advantage.

resistion
User Rank
Author
re: Memory-cube group issues initial draft specs
resistion   8/14/2012 11:47:31 PM
NO RATINGS
Agreed, so Intel not participating is important to wonder about.

rick merritt
User Rank
Author
re: Memory-cube group issues initial draft specs
rick merritt   8/15/2012 4:15:47 AM
NO RATINGS
Yes, Intel, AMD, HP and Oracle--basically the computer industry outside IBM--has been silent so far on its 3-D memory strategy. Presumably they are headed down the Jedec route.

David Brown
User Rank
Author
re: Memory-cube group issues initial draft specs
David Brown   8/16/2012 8:41:57 AM
NO RATINGS
Why would you want a DRAM controller in a processor? The DRAM controller handles things like timings for the different row selects, bank access, refresh, etc. You want these as close to the DRAM chip as possible. The reason memory controllers became integrated into processors is not because that's the best place for them - it was to avoid the extra chip, buses and latency caused by having the controller on a separate device between the processor and the memory (and sharing bus bandwidth with PCI and other buses). The reason the controller is not in the DRAM chips today is because the chip process needed for controller logic is very different from that needed for the DRAM cells, so it would be hugely expensive to put them on the same die. If you put the memory controller inside the DRAM cube, then the bus between the processor and the memory can be simpler and faster, and the memory controller can be more optimal for the dram banks it is controlling (including wider bus access and local cache inside the cube). The potential here is not just to increase bandwidth, but also to lower latency - especially in servers with large memories and ECC.

Wayman
User Rank
Author
re: Memory-cube group issues initial draft specs
Wayman   8/16/2012 6:36:23 AM
NO RATINGS
For Wide I/O, I suppose the target should be 12.8GBytes/s which is proposed by Intel.

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed