Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Engineers Build "Self-Healing" Chips Capable of Repairing Themselves

samzenpus posted about a year ago | from the electronics-heal-thyself dept.

Science 68

hypnosec writes "A Team of researchers and engineers at California Institute of Technology (Caltech) has developed 'self-healing' chips (PDF) that can heal themselves within a few microseconds. The team tested their work by damaging amplifiers in several places using high-powered lasers. In less than a second the chips were able to develop work-arounds thereby healing themselves."

cancel ×

68 comments

More accurate to say "More resilient chips"? (5, Insightful)

Looker_Device (2857489) | about a year ago | (#43137043)

Not to be too pedantic about it, but I'm very touchy about biological metaphors being inappropriately applied to technology (lets we forget how amazingly complex evolved biology really is compared to even our most advanced tech). FTFA, it sounds like they don't really "heal," they just reroute around the damage. But the damage is still there. It's more analogous to network packets being rerouted around a bad server than a biological entity actually replacing damaged cells.

Re:More accurate to say "More resilient chips"? (3, Interesting)

sensationull (889870) | about a year ago | (#43137141)

Thank you, that was what I was about to say, massively redundant, cool but it does not actually repair itself back to the way it was before, as it 'heals' it uses up that ability.

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43137221)

In a mission critical situation, lets say in space or medicine then you would want something like that, wouldn't you?

Re:More accurate to say "More resilient chips"? (1)

rioki (1328185) | about a year ago | (#43137445)

But that is a well understood problem on a macroscopic scale. The only thing they did war bring down the redundancy onto the chip. I don't think that it is that useful once a micrometeorite obliterates the entire chip.

Re:More accurate to say "More resilient chips"? (2)

bluefoxlucid (723572) | about a year ago | (#43137521)

Usefulness for a purpose has disconnected but relevant bearing on how it works. Operational details specify what problems the technology can solve; there is overlap for bare function, but the details are important. In this case you could have a chip with an inherent structure that would self-regrow when current is applied if damage is not extensive and materials were present (i.e. cracks self-heal infinitely, uses electricity); or a blue-goo type chip that contains a reservoir of consumable raw material to accomplish same purpose (consumes goo and electricity); or redundant schematics back-ups and FPGA (limited surface area to damage; potential routing problems); or redundant schematics with blue-goo (same, different implementation, in practice one will have an advantage of weight and effectiveness over the other but basic concerns are identical); or a simply redundant chip (prone to manufacturing deviations reducing its effectiveness; certain failures may be fatal; you need 100% additional capacity plus routing for each 100% redundancy; redundancy is per redundant component--100% blue-goo could repair the same 1% of surface through 100 failures, but 100% pure redundancy can recover from the same 1% surface failure 1 time).

This is not really "self-healing" but "redundant". Of the three above, you have these considerations:

Stable reflowing crystal: Can't heal from large damage; small damage (microcracks, minor fissures, electrical/mechanical stress) should heal. Chip normal lifecycle is extended indefinitely.

Blue-goo self-repair: Can heal from larger damage for a limited supply. Works off total damage: Eventual cumulative loss over 100% of chip area with 100% repair capacity will repair itself. Beyond capacity, repairs cease. Most likely, blue-goo would have transport issues into very tiny micro-fissures and cracks--cracks embedded deep inside substrate are not physically reachable and won't repair, yet may affect electrical properties of the substrate and thus impact chip performance.

Redundancy: Absolute healing from any type of failure. Redundancy takes up additional space for each redundant copy of each component: if you have 2 redundant copies of an ALU, you need space for 3 ALU total. If you lose one or two, it's functional. If you have one redundant copy of every component (100% coverage) and you lose two of any one, you lose the whole package at what may be 10% or 1% or more or less utilization. Contrast with blue-goo, where 10% loss means 10% loss of capacity to self-heal. Likely a high-stress area will require more specific redundancy; and bad luck could render a chip useless even with all that unutilized self-repair (redundant) capacity.

The holy grail would be a stable reflowing crystal with blue-goo that can repair the crystal. Tiny cracks would self-heal, while larger damage would trigger self-repair. Redundancy is worthless: if you have 3 ALU, you should be parallel processing instead of idling 2 of them for fail-over. Redundancy only seems desirable because we don't have a better method like blue-goo or stable reflowing crystal.

Re:More accurate to say "More resilient chips"? (2)

drinkypoo (153816) | about a year ago | (#43138153)

It seems dumb to me to have unused functional units lying around. But if a chip could detect that a functional unit has failed (by cross-testing?) and then degrade performance by not using it, while continuing to operate so that I can at least get useful fault information, that would be a massive win.

Re:More accurate to say "More resilient chips"? (1)

drkim (1559875) | about a year ago | (#43137565)

Thank you, that was what I was about to say, massively redundant, cool but it does not actually repair itself back to the way it was before, as it 'heals' it uses up that ability.

Not even new.

They have been building self-testing, redundant chips for years.

Here's a paper from 1982:
http://www.computer.org/csdl/trans/tc/1982/07/01676058.pdf [computer.org]
1988:
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&arnumber=3187 [ieee.org]

etc...

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43149353)

Thank you. I thought that I had seen this before(years ago), but couldn't quite recall where(and no your citations aren't the ones that I was thinking of either), so once again /. for the trendily titled "research" that has pretty well been accomplished before, but that pretty much sums up today's research anyways... no real breakthroughs, just evolution of prior research which for some reason ends up being bandied about like it was the greatest thing since sliced bread or something.

Re:More accurate to say "More resilient chips"? (1)

drkim (1559875) | about a year ago | (#43153011)

Thank you. I thought that I had seen this before(years ago), but couldn't quite recall where

To be honest, I was impressed with the first redundant chips (decades ago.)

Obviously, someone very bright said, "We're stamping these circuits on the die; let's put in a self-test, and just stamp a whole bunch." Even the Apollo missions back in the "60s had redundant CPUs (albeit not single chips) that would 'vote' on decisions, and vote out the odd man.

I just can't figure out why this is 'news'...
Ah /.

Re:More accurate to say "More resilient chips"? (1)

ikaruga (2725453) | about a year ago | (#43137729)

+1 if I had mod points. That was exactly my suspicion when I saw this news at other tech blogs. Reading the article a couple of days ago when I heard the news on Engadget, I was incredibly disappointed. The word "regeneration" is being completely misused here, unless that is it's meaning in Electronics(Just like "Teleportation" through quantum entanglement is quite different from Teleportation as we usually imagine). However as a System/Medical engineer that deals a lot of electronics I've never heard of such terminology. That is just a IC with some redundant reprogrammable capabilities that reprograms itself based on the input from the damage sensors it has built in itself. Nothing really goes back to the previous undamaged state. And after a number trials it will die for good. Useful tech, but misleading title.
Also horrible abstract. Just specs and numbers, exactly the opposite we expect from an abstract.

Re:More accurate to say "More resilient chips"? (2)

N!k0N (883435) | about a year ago | (#43137175)

Evolution's had a damn long time to get the "rebuilding cells" part down -- we're just at the "stop the bleeding" phase with the chips. Once they can rebuild their structure, we're in for trouble...

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43137261)

Evolution's had a damn long time to get the "rebuilding cells" part down -- we're just at the "stop the bleeding" phase with the chips.

Once they can rebuild their structure, we're in for trouble...

If I caught that hint correctly, what you're saying is we still haven't learned a damn thing from Cyberdyne Systems, and we actually want this tech.

Then again, why am I even worried. Religion will destroy us long before a model 101 will get the chance.

Re:More accurate to say "More resilient chips"? (1)

N!k0N (883435) | about a year ago | (#43137427)

Yeah, that's what it sounds like, the chips "heal" in the same way that networks "heal" -- route around the slow/bad/dead parts -- rather than biological healing of replacing the dead/missing cells. I'm taking this to be the first steps towards artificial healing -- the chips (or networks for that matter) can close off the parts that are "bleeding" due to damage.

So, for now the chips are able to put up a rudimentary scab. Eventually, they may be able to take "local" resources (silicon, carbon, whatever) and start rebuilding the patterns that were on them. I just hope the re-structuring there ends up with a "#5 is alive!" machine, rather than a T-1000.

Re:More accurate to say "More resilient chips"? (-1)

Anonymous Coward | about a year ago | (#43137481)

Then again, why am I even worried. Religion will destroy us long before a model 101 will get the chance.

Nice troll, asshole. You just insulted 3/4th of the people on the planet.

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43138899)

3/4th is actually fairly impressive. I doubt racism and sexism could get you over 1/2.

Re:More accurate to say "More resilient chips"? (-1)

Anonymous Coward | about a year ago | (#43137911)

being anti-religion, and agnostic are entirely two different things.

anti-religion is a religion.

Re:More accurate to say "More resilient chips"? (1)

Threni (635302) | about a year ago | (#43137195)

> they don't really "heal," they just reroute around the damage

But it says it in the title! Twice! They are 'self healing' AND they repair themselves! It leaves no doubt!

Re:More accurate to say "More resilient chips"? (1)

bill_mcgonigle (4333) | about a year ago | (#43137617)

But it says it in the title! Twice!

But that sounded better.

Re:More accurate to say "More resilient chips"? (1)

SuricouRaven (1897204) | about a year ago | (#43137255)

Does that make the broken-and-disconnected circuits scar tissue?

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43137337)

I agree, rather than "self-healing" it would be "self repairing". You don't heal a machine, you repair it. methinks there's way too much anthropomorphising these days.

If I hit it with a hammer, how well will it "heal"?

Re:More accurate to say "More resilient chips"? (1)

bill_mcgonigle (4333) | about a year ago | (#43137637)

You don't heal a machine, you repair it. methinks there's way too much anthropomorphising these days.

Yes, and it needs to stop because the language hates that.

Re:More accurate to say "More resilient chips"? (2)

jones_supa (887896) | about a year ago | (#43137931)

Not to be too pedantic about it, but I'm very touchy about biological metaphors being inappropriately applied to technology (lets we forget how amazingly complex evolved biology really is compared to even our most advanced tech). FTFA, it sounds like they don't really "heal," they just reroute around the damage.

Some of the biological processes also route around the damage, the brain being a good example.

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43138555)

Depends upon the severity, considering it's only been about 10 years since we realized that neurogenesis is real, I'd say that it's more likely the case that minor damage does get repaired, it's just that we don't normally notice it.

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43139265)

This is the same behavior seen in neurons, when a section of the brain is damaged. Replacing a cell would be analogous to having self replicating nano-machines inside the chip.

Re:More accurate to say "More resilient chips"? (0)

Anonymous Coward | about a year ago | (#43139687)

The word "ecosystem" is consultant-speak for "I'm full of shit"

I'm a limnologist, you clod.

Re:More accurate to say "More resilient chips"? (1)

TheEffigy (2666397) | about a year ago | (#43142379)

Not to be too pedantic about it, but I'm very touchy about biological metaphors being inappropriately applied to technology (lets we forget how amazingly complex evolved biology really is compared to even our most advanced tech). FTFA, it sounds like they don't really "heal," they just reroute around the damage. But the damage is still there. It's more analogous to network packets being rerouted around a bad server than a biological entity actually replacing damaged cells.

The brain is known to reroute signals in order to restore lost functionality in stroke victims, so (without having read TFA) I would group this under healing.

I for one... (1)

Anonymous Coward | about a year ago | (#43137063)

Oh never mind, it's just getting too easy nowadays.

Re:I for one... (0)

webmistressrachel (903577) | about a year ago | (#43137215)

How ironic that despite you and I noticing how redundant that meme is getting on here, the "First Post" (below yours right now) quotes what you were _going_ to say, precisely!

Re:I for one... (1)

jones_supa (887896) | about a year ago | (#43137941)

Or maybe it was the same AC who just couldn't resist the urge. ;)

First post (0)

Anonymous Coward | about a year ago | (#43137087)

I for one welcome our self-healing overlords

"A Team" (1)

Alopex (1973486) | about a year ago | (#43137129)

"If you have a problem, if no one else can help, and if you can find them, maybe you can hire the A-Team."

That BS again.... (4, Insightful)

gweihir (88907) | about a year ago | (#43137181)

They are NOT "self-healing". That would mean they can get back to their original state after damage. What these things have is a high level of redundancy. But whenever they suffer damage, the redundancy gets less and eventually they fail. Calling this "self-healing" is a direct lie.

Re:That BS again.... (0)

Anonymous Coward | about a year ago | (#43137289)

A dead transistor is a dead transistor. You can't repair the building blocks in electronics.

Re:That BS again.... (4, Insightful)

webmistressrachel (903577) | about a year ago | (#43137369)

| You can't repair the building blocks in electronics.

Yet.

Re:That BS again.... (2)

ledow (319597) | about a year ago | (#43137763)

Agreed.

But still has interesting implications for, say, radiation-hardened hardware like space-travel. Of course, it's nothing they don't already have in terms of the overall process, but having it on-chip is yet-another factor that has to experience corruption before you need to replace the hardware.

Another nice step, but nothing miraculous.

Re:That BS again.... (0)

Anonymous Coward | about a year ago | (#43138373)

"healing" even with respect to biological machines does not imply reverting to original state. a cut still leaves a scab. It simply refers to a restoration of function, often, but not always, to a lesser state.

Re:That BS again.... (1)

illestov (945762) | about a year ago | (#43140797)

I personally don't see a problem with calling a chip "self healing" if its capable of regaining its functionality through some sort of automated process. If you guys are going to be picky, then even in nature nothing is REALLY self healing. When a cell in your body gets damaged, it is not "repaired" but is replaced by another cell that performs the same function. When you cut your skin, you replace damaged cells with new ones and form a scar.

Re:That BS again.... (1)

gweihir (88907) | about a year ago | (#43142555)

The thing is that "healing" grows something new to replace what was damaged. This thing does not do that and all the spares have the same risk of getting damaged.

Re:That BS again.... (1)

illestov (945762) | about a year ago | (#43142793)

I would have to disagree with your definition..

heal (hl)

v. healed, healing, heals v.tr. 1. To restore to health or soundness; cure. 2. To set right; repair: healed the rift between us. 3. To restore (a person) to spiritual wholeness.

there is nothing in the definition that implies the process by which it heals. I think you are missing a much more interesting implication of that article which is that an IC that can diagnose itself and then switch to an appropriate "spare" or re-route itself is pretty amazing and can lead to a whole new family of micro electronics that is extremely reliable and perhaps eventually to .. Borg. ;-)

Re:That BS again.... (1)

gweihir (88907) | about a year ago | (#43146207)

However the definition implies restoration/repair of the defect. The thing from the OP just plugs in a spare and leaves the original broken. The only difference to component replacement is that the spare is already on the chip, and hence there is a hard limit on how often it can be done.

I also do not overlook the approach: It is pretty old, and there are reports of it from time to time. This is, at best, incremental research. Things like master-checker pairs of CPUs with some fail-over mechanism are well-known, but only in a limited community, as they are expensive.

Re:That BS again.... (1)

illestov (945762) | about a year ago | (#43146397)

haha ok you win. you sure know how to take excitement out of progress ;-)

Re:That BS again.... (1)

gweihir (88907) | about a year ago | (#43146661)

Well, excitement is nice when there actually is something to get excited about. It it is just marketing BS blowing things out of proportion, I like to try to be the voice of reason. Sorry about that ;-)

Will these heal up? (0)

Anonymous Coward | about a year ago | (#43137201)

... That's good. Because if you can't pass for human, you won't be much good to us.

another non-story (0)

Sterculius (2856655) | about a year ago | (#43137257)

I saw IBM demonstrate this with redundant hard drives about 15 years ago. They ripped out a failing drive (out of six) and the others "healed" the database without even much of a pause in availability. Then they popped in a new drive and the data was redistributed. So now a chip with built in redundancy can bypass damage, but without allowing for the bad section to be replaced. FAIL.

Re:another non-story (4, Funny)

fuzzyfuzzyfungus (1223518) | about a year ago | (#43137365)

Man, it makes me sick that people haven't taken the obvious step of giving the intricate metal layers and zones of dopant concentration on a silicon wafer the same modularity as 3.5 inch HDDs with hot-swap connectors... Scientists are so lazy.

Heck, why do we get worked up about integrated circuits at all? I saw Bell Labs demonstrate the same concept with discrete transistors before 1950, and they were basically just ripping off vacuum tubes...

Re:another non-story (0)

Anonymous Coward | about a year ago | (#43138681)

Um WTF are you talking about? Nothing you said makes any sense.

Re:another non-story (0)

Anonymous Coward | about a year ago | (#43141905)

He's making fun of the OP suggesting this development is a 'non-story' because some old feature written in software using off the shelf products already existed and therefore should be obvious and trivial to add to complex integrated circuitry. Software bits are easy to copy and recover, therefore transistors and connections should be trivial to copy and restore too!

Re:another non-story (4, Funny)

bluefoxlucid (723572) | about a year ago | (#43137543)

RAID is only 15 years old? It came about in like 1998?

Re:another non-story (1)

h4rr4r (612664) | about a year ago | (#43137795)

Maybe 25 years ago that would have been interesting by 1998 RAID was well known. There would be no pause in availability at all.

Re:another non-story (1)

jones_supa (887896) | about a year ago | (#43138039)

So now a chip with built in redundancy can bypass damage, but without allowing for the bad section to be replaced. FAIL.

Fail? There's again some typical /. thinking: an invention can't be useful if it's not perfect in all ways, thus it's a complete fail.

This is still an invention that can add a nice amount of robustness to mission-critical chips.

Re:another non-story (0)

Anonymous Coward | about a year ago | (#43138241)

Very much like a new chip that can replicate itself into a completely separate functional chip is not impressive because unix has the 'cp' command for years.

Rerouting power (1)

Anonymous Coward | about a year ago | (#43137301)

We all know what this will lead to....

https://www.youtube.com/watch?v=kQnwmEkEito

What a waste of time (0)

Anonymous Coward | about a year ago | (#43137321)

Don't these fools realize we'll all be able to 3D print our own chips at home soon? Any time now.

Re:What a waste of time (1)

webmistressrachel (903577) | about a year ago | (#43137399)

Seriously, [Citation Needed].

This is one of the few occasions where the meme is actually the most logical response. Unless, of course, my sarcasm detector is malfunctioning again.

Re:What a waste of time (1)

ACE209 (1067276) | about a year ago | (#43149007)

Unless, of course, my sarcasm detector is malfunctioning again.

Don't worry, you can 3D print a new one soon.

Real Genius? (2)

luke923 (778953) | about a year ago | (#43137531)

After reading CalTech and high-powered lasers, I could only think of a ragtag team of students like Mitch Taylor, Chris Knight, and Lazlo Hollyfeld implanting a two-way transceiver into Kent's dental work in order to thwart Hathaway's plans to embezzle funds from the DoD.

Sinclair? (0)

Anonymous Coward | about a year ago | (#43137841)

Didn't Sir Clive do this with memory back in the 80s?

Can you say? (0)

Anonymous Coward | about a year ago | (#43137977)

Nimrod? hehe

1 megabit RAM chip, 1980's (0)

michaelmalak (91262) | about a year ago | (#43137991)

My recollection is that back in the 1980's, chip manufacturers could not figure out how to make a 1-megabit RAM chip (that's 128KB, or a millionth the RAM of a server today) with no bad bits, so they added extra rows and the first time (?) it was utilized it would figure out which rows worked. For some reason, I recall that AT&T got a patent on it.

Can you say? (1)

Mr Mister (2862913) | about a year ago | (#43137997)

Nimrod !!!

Obviously engineers don't watch movies (1)

TheSkepticalOptimist (898384) | about a year ago | (#43138107)

Maybe watching the Terminator and Matrix movies might stop this kind of scientific "discovery".

Re:Obviously engineers don't watch movies (0)

Anonymous Coward | about a year ago | (#43145211)

Or Stargate SG1, remember those replicators?

Yay! (0)

Anonymous Coward | about a year ago | (#43138521)

I can't stand all the chip crumbs in the bottom of the bag. Self-healing chips to the rescue!

Ohhhh , , , (1)

Kimomaru (2579489) | about a year ago | (#43139029)

Thank god. I was worried we'd never get around to building Skynet.

Terminator (0)

Anonymous Coward | about a year ago | (#43139841)

Okay, just be careful about how sophisticated you make electronics and robotics. I've seen so many great breakthroughs lately. Robots can walk in a human-link manner and prevent themselves from falling while hurling extremely heavy blocks. Now, you can give them electronics that heal themselves? This is all fine as long as you don't give the thing real intelligence. I know researchers are working on that too. You put it all together and we will have real live Terminators on our hands.

Just be careful people....careful about the whole "human extinction" thing :)

Everything Old is .... (0)

Anonymous Coward | about a year ago | (#43142057)

Remember hearing about this concept a long time ago as a technique for spacecraft.

http://history.nasa.gov/computers/Ch5-5.html

Why is this news? (0)

Anonymous Coward | about a year ago | (#43142673)

I wonder why this is considered news. Routing around errors in many-core chips exists for some time now, and the only problem with doing it more is the cost of specialized hardware. Calling it "healing" is yet another publicity stint, and this sort of thing should not be rewarded.

OH THANK GOD! its about time (1)

Osgeld (1900440) | about a year ago | (#43144719)

My chips are always being damaged in specific places using high-powered lasers, and not the whole thing going up in a puff of smoke and small explosion if there is enough current.

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...