Caltech Develops Self-Healing System for Integrated Chips

Status
Not open for further replies.

dalethepcman

Distinguished
Jul 1, 2010
1,636
0
19,860
This is appears to be more of a "smart grid" approach, than "self healing" as this chip doesn't actually repair itself, it just attempts to route around a broken part to another part that can serve the same function.
 

danwat1234

Distinguished
Jun 13, 2008
1,395
0
19,310
Do modern microprocessors x86/arm have any redundancy if a few transistors fail, or some sort of checking and rerouting to disable a 'unit' if something is clobbered in there or a section of SRAM, etc? Or no?
 

azgard

Distinguished
Dec 20, 2002
159
0
18,680
[citation][nom]danwat1234[/nom]Do modern microprocessors x86/arm have any redundancy if a few transistors fail, or some sort of checking and rerouting to disable a 'unit' if something is clobbered in there or a section of SRAM, etc? Or no?[/citation]

A lot of designs incorporate tolerance provisioning, classic example are with GPU manufacturer's and binning of chip's as well as disabling and/or underclocking cores.
 

Vorador2

Distinguished
Jun 26, 2007
472
12
18,785
More than self-healing, it's CRC&ECC on steroids. In any case, for the space program this is really good news, since this allows a far greater degree of redundancy on electronics.
 

Non-Euclidean

Distinguished
Nov 5, 2009
463
0
18,810
[citation][nom]Abion47[/nom]I for one welcome our new Terminator overlords, just so long as they let me go back in time and get some Twinkies.[/citation]
No need, friend.

On March 12, 2013 it was reported, Twinkies are slated to return to store shelves by Summer 2013 after they, along with other famed Hostess Brands, were purchased out of bankruptcy by Apollo Global Management and Metropoulos & Co.
 

kenyee

Distinguished
Nov 17, 2006
176
0
18,680
Sounds like the custom ASIC inside the chip is a single failure point. Nice that they can reroute parts of the rest though...
 

gm0n3y

Distinguished
Mar 13, 2006
3,441
0
20,780
[citation][nom]kenyee[/nom]Sounds like the custom ASIC inside the chip is a single failure point. Nice that they can reroute parts of the rest though...[/citation]
Unless they build multiples ASIC units that can detect errors on one another. Of course if a unit itself is defective it might erroneously start deactivating other parts of the chip. Redundancy on the redundancy.
 

Duckhunt

Honorable
Sep 22, 2012
339
0
10,810
Damn fantastic. Will it be profitable?

What if they hit the power supply pin with the laser? or ground circuit attached to the pin for the ground? Can It use another pin to reroute?
 
Status
Not open for further replies.

TRENDING THREADS