In the Forums...
Posted: June 26, 2000
Written by: Tuan "Solace" Nguyen
PRISM and Holographic Materials
For a hologram to work in this technology, the material (holographic medium) must meet certain requirements.
Excellent optical quality: A high resolution data page with as many as a million pixels encoding digital data must be imaged through the material and onto the detector array, pixel for pixel. This requires very good homogeneity, and optical quality surfaces.
High recording fidelity: The material must faithfully record the data beam amplitude so that this high quality image can be reconstructed when the data is read out.
High dynamic range: The larger the number of holograms that are recorded in a common volume of material, the weaker each hologram becomes; the signal strength scales as the inverse square of the number of holograms, and is limited ultimately by the ability of the material to respond to optical exposure with the refractive index modulation that records the holograms. The greater is the materials ability to respond, i.e. the greater its dynamic range, the more holograms can be recorded, and ultimately, the greater the density of data that can be stored.
Low scattered light: The ultimate lower limit to the strength of holograms that are useful for data storage is determined by noise from readout beam scattering. Thus, scattered light also limits storage density.
High sensitivity: To store data in the material at a reasonable data rate, the material should respond to the recording beams with high sensitivity.
Non volatile storage: The material should retain the stored hologram for a time consistent with a data storage application, and should do so in the presence of the light beams used to read the data. For write-once read-many storage, an irreversible material (such as a photopolymer) can be used, which provides stable recording once exposed. If a reversible material is chosen in order to implement erasable/re-writable data storage, the requirement for nonvolatility is in conflict with that for high sensitivity unless a nonlinear writing scheme, such as two-color gated recording is used.
IBM estimates that it will take a few more years to tweak and refine the technology enough to build small desktop HDSS units. It guesses that desktop HDSS units could be available by as early as 2003. Okay, obviously I was joking about my HoloCube if you havenít noticed... heh. :)
HDSS hardware uses an acoustoptical (accu-whaa?) light deflector. This means a crystal whose refractive properties change according to sound waves traveling through it. To modify the beam angle, IBM thinks that an HDSS system can retrieve adjacent data pages in fewer than 100 microseconds. "Any convention al optical or magnetic storage unit will require some sort of mechanical means to access different data tracks, which takes on the order of milliseconds to accomplish," they say. "A gigabit-per-second data rate appears reasonable for holographic storage, and this should make it a cost-competitive leader with whatever exists."
Below is a diagram showing the core data page recording process of the technology:
Notice how the single beam is split into two.
Well folks, there you have it, the inside details of how this promising technology works. Personally, I canít wait until this technology arrives on the street, I wouldnít keep having to burn my uh, "backed up"... ISO images on to CDs. Anyhow, I hope youíve learned something from this article. Feel free to e-mail me if you have any questions or comments.
This isnít the end of the line for storage technology though. The next installment of my How It works series will cover something very interesting -- Atomic Resolution Storage. Until then.
References: Scientific American, Byte Resources, IBM Almaden Research Center, The Holography Handbook by Fred Unterseher and co-authors Ross Books 1987, The Complete Hologram Book by J. E. Kasper & S. A. Feller, Prentice-Hall, 1987, and Understanding Lasers, by Jeff Hecht, H. W. Sams and Co., 1988.
Click here for a printer friendly version.