Optics takes the computational edge

Optical computing has come a long way since the invention of the laser sources in the 1960's. Research has produced numerous architectures and applications, covering almost every aspect of computing. Some of these were feasible and became successful -- such as analog correlators and optical memories -- while others remained too esoteric or elusive to be widely adopted.

Applications using optical computing have been preffered to overcome the problems in the limited field space of old electronic architecture. There are quite a lot of reasons which have led to this step, firstly that light beams can propagate very close to each other, and even intersect, without any observable or measurable cross talk. This leads to the building of the dense array of interconnects using optical system. Secondly light is immune to electromagnetic interference,  which reduces the risk of noise even further. Thirdly,  light travels fast -- faster than any thing and it has extremely high band width. Moreover light can be harnessed even further for data processing.

The very next question which is confronted by most of us is, "How do we write information in to or off of a beam of light?" Well the answer is pretty simple, "by electrons." Sometimes the electrons are overt -- modulators , detectors and so on. Quite often they are covert, that is, nonlinear optics in which a material's optical properties are effected by charge carriiers, atoms, and so on interacting with the same or another beam of light

Goals of optical computing

Among the various goals of optical computing, the primary motive is "to make computing work better using optics." In the past considerable time, money and talent has been wasted just to compare why the optics should be given more emphasis as compared to the electronics. Well to answer it by an unbiased attitude, let me say that both optics and electronics have important roles but they are not the same. Our goal is to determine which technology is best for which purpose. Electronics is far maturer, far cheaper, far better funded.  The hope for optics lies in doing things provably impossible for electronics. Presuming that such tasks exists and are worthwhile, optics plays an essential role. If you want to do such a task, you must use optics.

To define what roles optics is best at, its perhaps easier to start by examining the areas in which optics isn't better than electronics.  We can then work toward areas in which it excels. By definition, digital computers can not handle continuous data. We once thought that it could approximate continuous computations arbitrarily well. The study of chaos shows, however, that we cannot be certain of. Optics, as defined, is itself continuous. The power consumption problem is also of great interest. Each digital computation takes a very small amount of engery, but if the number of computation is high, digital computers consume a huge amount of power. So only optics can implement massive, parallel, arbitrary mapping from an NxN output array using N weighted inter connections. Such a function can serve as the back bone of a truly massively parrallel neutral network or fuzzy programmable logic array. It can also implement any 2D - to - 2D mapping, whether or not that mapping is one to one. Optics can handle N on the order of 256 now, to provide parrallel, weighted interconnections. Providing such connections electronically would lead to an absurd mess of wires; we must therefore use optics over here.

The question still remains: What will a computer issue of a leading journal on optical computing look like in 5 or 10 years? We hope that it will include reports on the commercial success of free space interconnect - based systems, on the wide spread adoption of optical memories for massive storage in support of multimedia application (including video on demand), and advances made in micro - opto mechanical systems.

Holographic Data Storage

Of interest: Magnetic Storage and Recording: Materials and Devices by Asif Javed.

Computer users' hard disk drives are perpetually flowing and brimming up with data, even though only a year earlier the same size disk seemed more than sufficient. Research into the development of data storage devices is a race to keep up with this continuing demand for more capacity, higher density , and faster readout rates. Improvements in conventional memory technology -- magnetic hard disk drives, optical disks, and semiconductor memories -- have managed to keep pace wiith thedemand for bigger, faster memories. However, strong evidence indicates that these two-dimensional surface storage technologies are approaching fundamental limits that may be difficult to overcome, such as the wavelength of the light and the thermal stability of stored bits. An alterantive approach for next generation memories is to store data in three dimensions.

A hologram is a recording of the optical interference pattern that forms at the intersection of the two coherent optical beams. Typically, light from a single laser is split into two paths, the SIGNAL PATH and the REFERENCE PATH. The beam that propagates along the signal path carries information , where as the reference is designed to be simple to reproduce. A common reference beam is a plane wave; a light beam that propagates without converging or diverging .the two paths are overlapped on the holographic medium and the interference pattern between the two beams is recorded. Researchers achieve high density by super imposing many holograms within the same volume of recording material. Data is encoded and retrieved as two - dimensional pixelated images , with each pixel representing a bit . the inherent parallelism enables fast readouts rates: if thousands of holograms can be retrieved each second, with a million pixels in each, then the output data rates can reach 1 G bits per second.

HOLOGRAPHIC 3-D DISKS

Among the various approaches an approach to spatial multiplexing leaves the optics and components stationary and moves the storage material. The simplest method to do this employs a disk configuration. The disk is constructed with a thin layer of the holographic material. Multiple holograms can be stored at each location on the disk surface. These locations are arranged along radial tracks, with the motion of the head selecting a track. Typically , it would be desirable to fabricate disks of 1- mm thicknesss, yeilding a density of approximately 100 bits per squared micron. Even though the data is still stored in 3D, in the disk configuration the surface density, not the volume density, is what matters for most practical purposes.

Holographic storage is a promising candidate for next generaation storage. Recent research has demonstrated that holographic storage
systems with desirable properties can be engineered. The next step to build these systems at costs competitive with those of existing
technologies and to optimize the storage media. If suitable recording materials become available from the research efforts currently under way, we envision a significant role for holographic storage.