New Optical Chip Claims 8 Trillion Operations/sec.

Richard Finney writes "Lenslet is announcing the 'World's First Commercial Optical Processor.'. Reuters has the story here. The Inquirer has a cool graphic here on it. The processor is specified to run at a speed of 8 Tera (8,000 Giga) operations per second, one thousand times faster than any known DSP. When Lenslet releases its Enlight processor in a matter of weeks, a unit using the technology will be 1.7 centimetres high and measure 15 by 15 centimetres."

Comments

The Definitive Question Is: +1, Patriotic

[Anonymous Coward]

But will it run Windoze?

This is the Future

[Doesn't_Comment_Code]

I've done a lot of research on this. Optical processors have incredible potential. And if you think that's good, just wait. The combo of an optical processor with optical memory is a one-two punch. This is definitely the future of computing.

Re:This is the Future

[back_pages]

For real, what's the point of 8 trillion operations per second when there's no existing memory to support it (of which I am aware)? So this chip runs REALLY REALLY FAST on code that's REALLY REALLY SMALL, and otherwise it's bottlenecked by the memory bus and memory speeds.

I appreciate that it's a great demonstration of new technology, but maybe it's a little premature to call this a new commercial chip. It sounds to me like a demonstration of a research project or an exposition of things to come.

It's quite possible that I'm completely ignorant about this, but to whom do they expect to sell the latest and greatest THREE ORDERS OF MAGNITUDE increase in memory bottleneck?

Re:This is the Future

[Bendebecker]

What I meant was a computer that used both electricity and light. From what I understand, the first electric computer used electric relays like the ones used in phones (cheaper than vaccum tubes whch had only been invented the same years as the germans began building the machine) and was described as being an electronic computer. When ENIAC was built using vacuum tubes, it was described as being a digital computer. What ever word you want to use to designate that the computer functions using electricity and transistors instead of light is up to you at this point.

I'm not being funny...

[Pingular]

would you be able to link this in a Beowulf-type manner?

FYI

[Doesn't_Comment_Code]

Interstingly, optical processors aren't faster because light is faster than electricity. They are faster because they have much faster rise and fall times between digital on and digital off.

Re:FYI

[QuantumFTL]

Interstingly, optical processors aren't faster because light is faster than electricity. They are faster because they have much faster rise and fall times between digital on and digital off.

While your statement may indeed be correct, that is not why this chip is faster. The reason is that they are doing analog signal processing using the physics of various optical elements to perform computationally intensive mathematics.

Think of it this way: We can use large, expensive mathematical operations to simulate optical components, which means we can also do the reverse - using optical components to perform the expensive mathematical operations.

I read about this about 2 years ago, and it was really quite fascinating to me. It turns out that with a simple lense, you can compute a fourier transform just by focussing the light (it doesn't focus down to an infinitesimally small point).

I managed to find an article about this, hopefully it should be apparent why this chip doesn't run quake:
Check it out here.

They are certainly not the only people doing this. I've seen plenty of references of this being used in missile guidance systems (turns out a simple fourier transform trick can be used to track objects in a camera). Someone I met while working at the Jet Propulsion Lab was working on this Optical Signal Processors. They prove to be very big in the next 10 years.

Cheers,
Justin

Environment

[Whispers_in_the_dark]

What sort of environment would this sort of device need in order to operate? Glancing at the picture I looks like the device internals need to be very precisely aligned to work. How does it react to vibration? Temperature? Phase of the moon (kidding)? Would a regular CEV style environment be sufficent or does it require uber-protection?

Just curious...

Re:Why the defense industry concentration?

[Zakabog]

Ummm because the company is in Israel, a country that has to deal with terrorist type attacks on a daily basis? I thought the same question till I saw

"...said Major-General (Ret.) Isaac Ben- Israel, former head of the R&D Directorate of the Israeli Ministry of Defense."

What else is the former head of the R&D Directorate of the Israeli Ministry of Defense going to say about a new chip like this one?

Re:User availability...

[stratjakt]

First off, this thing is going to cost a bazillion dollars, and will be a "commercial beta" product for years.. They'll sell them, but theres still a ton of work to improve the thing.

So who would buy one? Someone with lots of cash - the DoD.. You wanna sell to the DoD, you have to show how it will fight terrorism.. It's just the way things are selling to the government.

I work writing code for public safety systems - records and dispatching for cops. We did a search and replace on all our marketing literature, replacing "gang member" with "terrorist".

I don't like seing "terrorist" being morphed into the buzzword du jour, it lessens its value to the point the word will be meaningless soon enough. Like "racist". The word "racist" has been thrown around so much that it doesnt mean anything. Bush is a racist, Clinton was a racist, Jesse Jackson is racist, Phil down the street is racist, N'Sync - all racists. Bleah..

But thats just how politics work.

Memory is irrelevant for this kind of "processor"

[wowbagger]

This is NOT a Harvard architecture part - this isn't fetching instructions from RAM and executing them, like a regular DSP would.

Think of this more like an FPGA - you have a device that is configured for a specific processing algorithm, and data is fed in at wire rate and processed at wire rate.

An example of how a device like this might be used may be in order:
I'm trying to find a radar pulse buried in the noise coming in from my receiver. I want to know the phase delay of the radar pulse - how long from when I sent it till I got it back.

Now, I know what my radar pulse looks like as it goes out. I know that any reflection is going to consist of versions of that pulse shape, delayed and of varying strengths. So what I do is called a correlation - the easiest way to think of this is to imagine having 2 transparencies, one of my outgoing pulse, and one of the incoming signal. Now, I hold them up to the light, and slide the incoming signal across the reference pulse until things match up - that's the point of maximum correlation, and that give me the delay of the signal.

A real correlation function is a bit more complicated as you have to allow for the signal level to be changed - if I am looking for a signal of N samples in a received data stream of M samples, I have to do M*N multiply and add operations to get my correlation. Now, for a radar signal I might be sampling at over a billion samples a second, and looking for a chirp of a 100 ns would give me over 100 billion MAC operations a second. There are ways to do that with conventional DSPs, but they are a galloping BITCH to do (you basically make a cluster of DSPs, and each DSP takes a part of the signal. Synchronising that is a bitch.)

This device would work by having the shape of the outbound pulse represented in the structure of the device itself, and the MACs are done by taking the incoming data stream and projecting it on the structure - thus you do all your processing in parallel, and at wire speed. You get a pulse out when the incoming signal matched the signal you ar looking for.

Window of opportunity:The Open Source Optical RISC

[NZheretic]

The introduction of this technology over the next five or so years offers a window of opportunity to collectivly develop Free/Liberty/Open Licensed optical RISC ( Reduced Instruction Set CPU) technology.

The shift from electronic to optical results in a massive reduction in the time it take to change states, to the point where it possible to, once again, build a CPU from relatively widely spaced modular optical components. You can build a single optical CPU spread over a motherboard or even cabinet sized area and it will still be several magnitude times faster than the fastest silicon/electronic single chip CPU.

No one but the biggest companies are going to have the capital nessary to collect and shrink the resulting designs down into single optical chip hardware and manufacture the result, with a further magnitude increase in performance. As with the existing CPU industry, it is likely that the market could maintain only a few such CPU companies. Opening up the design and development process, as with open source development, would result is a far more rapid pace of development. Relative obsolescence woul;d insure that there would plenty of opertunity for large profits for the large and small manufactures.

Re:More Info

[Mozz Alimoz]

I'll have a guess at how this works.

The article says "The Ablaze(TM) is the Spatial Light Modulator (SLM) in the optical core of the EnLight256(TM)". Going by the graphic in the Inquirer article, they shine a row of blinking lights through a LCD-like device (and some lenses and mirrors I assume) and collect the results in a column of light sensors on the other end.

Each pattern of on/off elements on the LCD-like device gives them a different transformation running at however fast you could emit and sense the light. I doubt they mechanically move the optical arrangment so that would seem to limit the number of transformations. Some of the LCD patterns might give useful transformations. A vector multiply, a Fast Fourier Transform (maybe) or a sort (I doubt it)?

If the numbers are an analog light intensity level the precision would depend on how precise the light emitters and sensors you have are. Packaging the mirrors and lenses small enough is a neat trick. Having a problem that fits the available transformations and can supply data in and out fast enougth is another. I wonder anything useful can be done by quickly switching LCD matrix pattern, or directly feeding outputs back as inputs?

Re:Memory is irrelevant for this kind of "processo

[Anonymous Coward]

He's right. This dsp will really revolutionize how and what we are capable of doing. In terms of RF systems, we have hit an era where superheterodyne setups are no longer required for some (my) microwave work!! I'm an RF engineer at NASA and people here are shitting muffins after seeing this article. Esp since we were just given a bucket of money to build a variation on the thingamajig the parent mentioned and we couldnt find fast enough dsps ('clustering' not an option) to do our dirty work so we have to build a correlator out of high speed logic gates from scratch. Wheres 'add to cart' ?!?!

I ran one in 1967. B-)

[Ungrounded Lightning]

These things are NOT NEW. in 1985 I was a Jet Propulsion laboratory. A caltech professor there was using a light modulator to perform convolution matrix a operations to decode synthetic aperature radar data. THe design is identical.

I was a tech in Emmet Leith's "Radar and Optics" lab at the UofMich and one of the first things I did was run an optical processor using essentially this hack - again to process synthetic aperture radar data. This was in 1967.

Multi-megapixel 2-D FFT plus some geometry corrections in the time it took the laser light to go from the input film plane to the output film plane - about 6 feet on that device.

We were already considering how to replace the photographic film input and output devices with electronic substitutes in those days, too. The size of the device we used was large only because it was convenient to construct it with aluminum U-beams and stock lasers, lenses, and lens holders. Given decent I/O, making a disk-drive sized model, say to do realtime processing in an aircraft-mounted radar, would have been trivial. (The signals to be processed were already electronic and at reasonable bandwidth - lower than a TV image.)

Nowadays this is done by DSPs. Why? Because they're adequately fast and are FLEXIBLE. Optic processors do only one type of computation, and require physical adjustment to tune the parameters. If you can do that computation on something more general-purpose, as fast as your data arrives, why bother building something larger and more limited to do it faster?