Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Seeking the Fourth Circuit Element

F.Y. Teng | May 25, 2010
A conversation with the czar of R&D at HP Labs about taking memristors from paper to production to mobile computers to enterprise infrastructures.

What should the industry, media and more importantly your customers expect from you in the next 12 months? Perhaps, you can start by talking about your development-ready architectures for memory chips using memristors?

What people can expect in the next 12 months is more surprises out of the research lab. As I mentioned, it will be 18 months before we will have results that we would be interested in talking about from the development phase. So in the next 12 months, there certainly will be more interesting research results coming out because its like a rose thats opening up. As we study this memristor, we are learning more and more about its properties and what it can do and that is sort of an interesting phase. But it will be at least two years before we would start making any announcements of products. Theres no specific product being diskussed. Right now we need to get through a development stage first. I will, however, say that there is a major emphasis in HP to shorten the timeline between research and product to make that whole process as efficient as possible and make is as short as possible.

That may be frustrating for us and other people because it means that it does inject some uncertainty. But we dont have an absolutely finished thing right now that we can latch onto and say with absolute confidence that something is going to appear on such and such a month. We are literally still doing some research in parallel with development and then we will be doing that in parallel with trying to bring out a product. We are trying to go from cycle times of say 15 years from diskovery in the labs to first product, and cut that in half, perhaps.

What else is your Quantum Systems Lab doing right now?
Well, were working on a couple of other technologies that we find very interesting. The memristor is for storing information primarily right now, but we believe in the meantime if information moves more than a centimetre, it should move as photons rather than electrons and so one of our major projects is called Photonic Interconnect, which essentially means moving data around using light instead of electricity. This has been done for decades by telephone companies for telephone conversations. This [telephone] conversation right now is being carried out several thousand miles under the Pacific Ocean on a fibre optic cable, but that technology is much too expensive for computers so we are working on bringing the fibre optic revolution into computer systems by essentially creating integrated photonic circuits or integrated optical circuits that bring the cost of optics down to the point where we can use this power in a computer system.

 

Previous Page  1  2  3  4  5  6  Next Page 

Sign up for Computerworld eNewsletters.