Slashdot Mirror
Researchers Develop Photonic Processors
TheCybernator writes to mention a New Scientist story about scientists who are developing a light-based processor by actually storing and delaying photons. These 'optical buffers' may one day be used to make super-fast microchips based on light instead of electrons. From the article: "A decade from now ... there [may] be not seven cores but hundreds on a chip ... Connecting these cores using light could solve this problem. Until now, the lack of optical buffers has been a key roadblock to these kinds of light connections. The way information is transmitted means that buffers must hold packets of data while a router decides where they are to be sent. Buffers are also needed to delay optical pulses - so they do not collide at switching points - and to synchronise streams of data coming from different places."
Something like 20 years ago, I heard about a device that used interference in a crystal with the upshot that the presence of light on one facet of a crystal determined whether light could go through the crystal along another path. This was supposed to form the basis for entirely-optical logic devices. Does this ring a bell? Anyone?
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
If the ionization-rate is constant for all photonic entities, they can really bust some heads... in a optical sense of course.
I look at my process list. There's a hell of a lot of processes and threads. Even without any special programming for any given application, you still get the very handy benefit of less context switching. Maybe even 1 process per core :P I'll take that anyday.
Cool! I should still be alive then!
The world market for photonic computers will only be five or six.
Isn't that like saying "640KB ought to be enough for anyone"?
I can't speak for Windows or OS X, but Linux's SMP implementation locks a process to the CPU it first runs on unless it absolutely has to move it to distribute load evenly.
A deep unwavering belief is a sure sign you're missing something...
Aside from the mythical status of that statement, no. The logistical difficulties in threading that many cores and using them efficiently, especially in personal computing, is extreme. I have no doubt that at some point this will be done, but a number of breakthroughs in computer science and a paradigm shift in programming techniques is required first.
Slashdot: Playing Favorites Since 1997
Hi,
1) You can have 32--64 "cores" (depending on how you define it) on a Si ship from Sun now. I'm using a 24-thread T1000 now and it's great.
2) I assume this was a troll post, since there are many many many "embarrassingly parallel" scientific/financial/Web problems...
Rgds
Damon
http://m.earth.org.uk/
The important distinguision to make when comparing the benefits of going massivly paralel processing is that it is possible to solve NEW problems in realtime with these processors. Eg, we don't need to run Word 100 times faster, however we can get eg. games and scientific simulations (two sides to the same coin) that uses detailed physics engines and realtime raytracing. Raytracing can be almost naivly paralellised with up to as many processors as screen pixels. I remember using a computer with 65536 processors called the maspar which was built in the early 90's. Our main use for this computer was for image processing which also could easily be parallelized. It just took a bit of a shift of perspective to learn how to program it since it was SIMD (Single Instruction Multiple Data) but boy where it fast for it's time.
Physics is a bit more difficult but there are tehniques too for paralellization utilizing the fact that object interactions form islands of connected parts. Eg, when simulating your hair in a realistic way don't test for interactions with the objects in a distant part of the game. Physics engine are just starting to become used for these purposes but can easily require how much CPU power you want for it. Simulating eg. the clothes in the game characters or dynamic subdivision of parts as they break or bend due to forces (do you want realistic dents in your car after hitting that pedestrian?). These would both require an order of magnitude more CPU power than what we can do in realtime physics today.
So, to make a short summary. Yes, we can always achieve new tricks with even more computing power. Give me a cluster of a million processors and i would still complain that it's too slow for what i want to do.
Speed of electricity =! speed of light.
speed of light insite a metal =! c.
In copper lanes like on modern cpus, the speed is about 30-35% of c.
Photonic crystals and optical fibers, otoh, can have a permiability that allows speeds of near C for photons.
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
electricity is about a third the speed of light and is now one of the big bottlenecks in high performance computing.
ie, one of the reasons the cache on a processor is so much faster than going out to ram is that it is physically nearby. processors today move fast enough that several cycles idle while just waiting for ram to return data through the winding conductor path.
-
We are starting to get to the point where the capacitance of the tiny little wires in a genuine concern, and crosstalk between them is also significant. Also, the amount of space taken up by wiring is annoying. You can use a single waveguide with several frequencies of light to replace several wires and solve all those problems at one. At least, in theory. In practice, it's really hard to build it. But, it'll be pretty sweet when we get it all sorted out.
I guess the author didn't know 100+ core chips are already out, http://www.rapportincorporated.com/, with a 256 cored chip already for sale. They are already coming out with a 1024 cored chip. In fact, IBM has already entered a partnership with them creating a multichip, PowerPC core + 1024 cores for a 1025 cored chip.
Hopefully using light instead of electrons would cut down on the amount of heat that is dissipated. Otherwise, a 256-core processor could serve double duty as a furnace for a 3000 sq foot home in winter.
---------
There is inferior bacteria on the interior of your posterior.
OK. Now that the (semi-) joke is out of the way, a 100 processor core would have a ton of uses - large scale Monte-Carlo simulations (used in everything from AI, to biostatistics, to computational chemistry); verification of logic circuits, microcode and tests for both; large-scale optimization problems; high-speed rendering for scientific visualization and entertainment purposes; and the list goes on. Oh yeah, if you had more cores than processes, your computer might be a bit more responsive than it is now.
Of course, the problems mentioned above are some where computation speed is the bound and which are (relatively) easy to make parallel. Even so, the main obstacles to using this kind of circuitry would be optical memory size and ability to do enough I/O to keep the optical memory subsystem fed. This sort of I/O balancing is a problem, though, in all systems and should be easy enough to solve by balancing on-chip memory size against the number of cores.
That is all.