TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

'Human brain' supercomputer with 1M processors switched on for first time

93 pointsby theaeolistover 6 years ago

20 comments

llcoolvover 6 years ago
&quot;‘SpiNNaker’ machine is capable of completing more than 200 million million actions per second, with each of its chips having 100 million moving parts.&quot;<p>Could someone elaborate? I am probably missing something as I hadn&#x27;t heard of moving parts on a solid-state device?
评论 #18383373 未加载
PopePompousover 6 years ago
From the article: &quot;The newly formed million-processor-core ‘Spiking Neural Network Architecture’ or ‘SpiNNaker’ machine is capable of completing more than 200 million million actions per second, with each of its chips having 100 million moving parts.&quot;<p>I kinda doubt that.
评论 #18382873 未加载
评论 #18382952 未加载
评论 #18382851 未加载
评论 #18382879 未加载
YeGoblynQueenneover 6 years ago
Article author:<p>&gt;&gt; The world’s largest neuromorphic supercomputer designed and built to work in <i>the same way a human brain does</i><p>Project lead:<p>&gt;&gt; We’ve essentially created a machine that works <i>more like a brain</i> than a traditional computer<p>Press releases, ladies and gentlemen.
marmadukeover 6 years ago
I went to a code jam with some of these guys (as part of the Human Brain Project), the architecture is pretty interesting but it&#x27;s lots and lots of little ARM (v6?) processors on a grid interconnect, probably not too far from Xeon Phi, even if it aims for neural like computation more so that a Phi.
评论 #18383603 未加载
评论 #18384877 未加载
JoeAltmaierover 6 years ago
Does that add up? Doesn&#x27;t the human brain have several more zeroes on there? Maybe &#x27;worm brain&#x27; or &#x27;lizard brain&#x27; would be a better description?
评论 #18383198 未加载
评论 #18383227 未加载
peterlkover 6 years ago
This article was pretty low on details, so I went out and collected some links:<p>This looks more or less like another stab at the Cray connection machine[0], but with modern hardware and a better framework about how neural nets can and should work.<p>&gt; The SpiNNaker engine is a massively-parallel multi-core computing system. It will contain up to 1,036,800 ARM9 cores and 7Tbytes of RAM distributed throughout the system in 57K nodes, each node being a System-in-Package (SiP) containing 18 cores plus a 128Mbyte off-die SDRAM (Synchronous Dynamic Random Access Memory). Each core has associated with it 64Kbytes of data tightly-coupled memory (DTCM) and 32Kbytes of instruction tightly-coupled memory (ITCM). The cores have a variety of ways of communicating with each other and with the memory, the dominant of which is by packets. These are 5- or 9-byte (40- or 72-bit) quanta of information that are transmitted around the system under the aegis of a bespoke concurrent hardware routing system. [1]<p>So, lots of relatively relatively tiny, interconnected nodes.<p>They built their own SoC to handle this. With a built-in router in the middle. The router handles routing on the chip, and multicasts to its neighbors.<p>&gt; The heart of the communications infrastructure is a bespoke multicast router that is able to replicate packets where necessary to implement the multicast function associated with sending the same packet to several different destinations. [2]<p>It also looks like they&#x27;re developing dev boards [3]<p>So basically, this looks like a giant, really awesome, custom ARM cluster that they want to do neural network stuff with.<p>If anyone from the team is here, I&#x27;d love to hear more about how this will be used. Specifically, how will you prevent SpiNNaker from going down the same path as the Connection Machine - (stops doing AI stuff because, say, geneticists want to use it for protein sequencing)? Why do you see this as the future over something like NVIDIA&#x27;s new HGX-2 or clusters of TPUs?<p>[0] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Connection_Machine" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Connection_Machine</a><p>[1] <a href="http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;architecture&#x2F;" rel="nofollow">http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;architectu...</a><p>[2] <a href="http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;SpiNNchip&#x2F;" rel="nofollow">http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;SpiNNchip&#x2F;</a><p>[3] <a href="http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;hardware&#x2F;index3.php" rel="nofollow">http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;projects&#x2F;SpiNNaker&#x2F;hardware&#x2F;i...</a>
评论 #18385557 未加载
评论 #18384516 未加载
评论 #18384235 未加载
shshhdhsover 6 years ago
&gt; capable of completing more than 200 million million actions per second<p>&gt; To reach this point it has taken £15million in funding, 20 years in conception and over 10 years in construction, with the initial build starting way back in 2006.<p>Wow, those numbers.. and 10 years to build... I’d be very excited to turn it on!
评论 #18383027 未加载
评论 #18384332 未加载
EamonnMRover 6 years ago
&gt; Biological neurons are basic brain cells present in the nervous system that communicate primarily by emitting ‘spikes’ of pure electro-chemical energy.<p>I don&#x27;t think that those are terms of art.
nyrulezover 6 years ago
This article, while obviously PR, is confusing to me. It seems to pin too much emphasis on the nature of its hardware as if that&#x27;s enough to &quot;emulate&quot; the brain. If they don&#x27;t have folks at the calibre of Deepmind to drive this thing, can it really go very far?<p>What it could be useful for is neural structure modeling at a more primitive layer, even if the end outcome isn&#x27;t usable for practical consumption.
评论 #18385668 未加载
bra-ketover 6 years ago
Furber&#x27;s lab built an interesting extension on SDM with N-of-M rank codes <a href="http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;ftp&#x2F;pub&#x2F;apt&#x2F;papers&#x2F;sbf_TNN07_old.pdf" rel="nofollow">http:&#x2F;&#x2F;apt.cs.manchester.ac.uk&#x2F;ftp&#x2F;pub&#x2F;apt&#x2F;papers&#x2F;sbf_TNN07_...</a>
ytersover 6 years ago
Some are saying this is no where near the amount of processing the human brain has. But, it doesn&#x27;t seem we need that much of the brain. There are plenty of stories of highly functional and talented individuals with almost no brain, and this machine would probably be in their ballpark.
评论 #18384469 未加载
评论 #18384864 未加载
mellingover 6 years ago
Steve Furber did some early design work in the ARM processor:<p><a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Steve_Furber" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Steve_Furber</a>
drannexover 6 years ago
Happy to see Stanford R. Ovshinskys dreams being created, even thought it is years past his death.
mrecover 6 years ago
I&#x27;m getting NET::ERR_CERT_AUTHORITY_INVALID with Chrome - am I the only one?
评论 #18383419 未加载
qwerty456127over 6 years ago
And what did it say?
评论 #18383837 未加载
评论 #18384520 未加载
sakersover 6 years ago
Shortly after being switched on, researchers where found telling SpiNNaker exactly what it could do with the lifetime supply of chocolate.
MBCookover 6 years ago
Only time will tell if it’s a fair dinkum thinkum.
pixelbreakerover 6 years ago
Won&#x27;t be long before it&#x27;s smoking Spice.
ctdonathover 6 years ago
&quot;...and even before its data banks had been connected, it deduced the existence of rice pudding and income tax before anyone managed to turn it off.&quot; - H2G2
monk_e_boyover 6 years ago
1M ... 1,000,000<p>Powers of 10 ... 10 fingers on the ape-man. Such a weird non computing number to be thrilled with.<p>I&#x27;m always suspicious when numbers fit into powers of ten like that. Like, somewhere in that build process the person who holds the purse strings doesn&#x27;t know binary.
评论 #18383253 未加载
评论 #18385727 未加载
评论 #18383425 未加载