Breaking News
News & Analysis

Flash DIMMS Invade DDR3 Slots

7/30/2013 06:00 AM EDT
8 comments
NO RATINGS
1 saves
< Previous Page 2 / 2
More Related Links
View Comments: Newest First | Oldest First | Threaded View
rick merritt
User Rank
Author
Re: Will you plug flash into your memory bus?
rick merritt   8/6/2013 9:53:54 PM
NO RATINGS
@an_m: You got it, and I suspect that's just one of many implicaitons people will be discovering. There's a lot of new software to be written.

an_m
User Rank
Author
Re: Will you plug flash into your memory bus?
an_m   8/1/2013 2:48:02 PM
NO RATINGS
Hay.

  if we have the programs stored in flash on the same bus as the dimm,

     then we dont need to spend all that time moving executable blocks  from drive to memory to run them, execute in place.

'just' need the code to know how to execute in place and only write to the flash when needed,

now that WILL speed things up, 

sounds fantastic,

 

rick merritt
User Rank
Author
Re: Will you plug flash into your memory bus?
rick merritt   7/31/2013 10:30:49 AM
NO RATINGS
@ccvv: Yes it will be facinating to see which of the array of new contenders takes over--or maybe we will wind up with a much more diverse world in memories.

rick merritt
User Rank
Author
Re: Will you plug flash into your memory bus?
rick merritt   7/30/2013 5:24:25 PM
NO RATINGS
@Sanjob: Endurance of flsash relative to DRAM remains an issue though flash chips and their controllers have made significant progess in this area.

rick merritt
User Rank
Author
Micron link, Msoft is key
rick merritt   7/30/2013 5:23:25 PM
NO RATINGS
Here's the link to Micron's plans for NVDIMMs I could not finmd earlier:
2013-02-07Flash will ride DRAM bus in 2014, says Micron
Micron notes support under Windows is key for this approach


ccvv
User Rank
Author
Re: Will you plug flash into your memory bus?
ccvv   7/30/2013 1:37:58 PM
NO RATINGS
Newer class of NVMs are coming - as a matter of fact, they are already in production - such as Phase Change Memory (PCM) which take the cycling capability of NV memory into millions of cycles.

Sanjib.A
User Rank
Author
Re: Will you plug flash into your memory bus?
Sanjib.A   7/30/2013 12:02:53 PM
NO RATINGS
Don't know much about the NAND FLASH technology; But the first thought that occurs to me is the numbers of write/erase cycles that could be allowed for a NAND FLASH before it loses its capability for reliable operation. Is that number high enough as compared to DDR SDRAM? If FLASH is written/erased so frequently as like SDRAM how long it will last compared to a SDRAM?

Otherwise it will be great to have NV memory in place of the volatile memory...can get rid of batteries/super caps for implementing battery-backed memory.

rick merritt
User Rank
Author
Will you plug flash into your memory bus?
rick merritt   7/30/2013 11:13:27 AM
NO RATINGS
Why or why not?  Anyone else trying to enable this?

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