Printer Friendly

Storage technologies--the long view.

For vendors and end-users alike, the progress of technology marks a never-ending process to successfully store, retrieve, migrate and effectively manage the entire data life-cycle process. We have extensively focused on the management challenges in trade journals, conferences and the media in the past few years, recognizing that the biggest storage challenge is its management. In order to better understand how we are preparing our storage infrastructure for our future, we must take some time to look ahead and see just what it is that we may likely be managing.

Disk Futures Roadmap

As of mid-2003, just a few significant disk drive suppliers remain: Seagate, Maxtor, Western Digital, IBM, Fujitsu, Samsung, Toshiba, NEC and Hitachi. With Hitachi's purchase of IBM's disk drive business, the majority of the world's disk drive producers is now Asian based, although the majority of disk drives are still produced by North American-based companies Seagate, Maxtor and Western Digital.

Magnetic mass storage technology advances have enabled the migration of disk units to 3.5-inch and smaller diameter form factors. The 3.5-inch and 2.5-inch form factors are present in all segments of the market: server, desktop, and mobile. The recent introduction of a 1-inch form factor may point the way to future configurations of the disk drives though these presently remain confined to the entertainment sector of the storage industry. There are many signs that 2.5-inch form factor disk drives will soon start to displace 3.5-inch drives as the de-facto standard due to the increases in areal recording density.

Magnetic disk areal density has grown at an impressive 60% compound annual growth rate historically and has accelerated to greater than a 60% rate since 1999. It is likely that the rate of disk areal density increase in magnetic recording will start to drop below 100% annually due to the greater difficulty in making the new technology work. A particular technical challenge will be making magnetic recording heads with track width dimensions that are smaller than the minimum feature size of the optical lithographic equipment used in the semiconductor industry. From a usage perspective, the access density of increasingly higher capacity disks has become a mounting performance concern even though the price per megabyte steadily declines. Over the next five years, the likelihood of perpendicular recording using a patterned media may appear to further increase capacities. Perpendicular recording enables the bits to be magnetized on end, perpendicular to the disk surface (rather than parallel to the surface). This greatly improves recording density and the raw drive data rate. Thermo-mechanical data writing on plastic substrates continues to achieve positive results and may bypass the anticipated super-paramagnetic limit completely.

Tape Futures Roadmap

Although somewhat less form factor driven than disk, tape drives are also physically shrinking; many drive product offerings are now in the 5.25 -inch and 3.5-inch form factors. Most formats have settled on the 5.25-inch cartridge. The very low-end (single-user systems) of the tape market continues to shrink in the face of small form-factor removable disks. In the middle markets and enterprise sectors, automated tape libraries have become commonplace and range in capacity from tens to thousands of tapes, with total capacities now exceeding I petabyte (1PB = 1015 bytes) per library. Tape cartridge roadmaps identify native capacities up to 8 terabytes and drive data rates up to 800 megabytes per second.

Magnetic tape continues to offer the highest probability of success in achieving the maximum total volumetric density and maximum data rate combined with the lowest cost per megabyte of any technology. The simplified use of tape has made tape accepted in a wide variety of data storage applications and tape is no longer only positioned as a backup/recovery technology. In the interactive storage environment, this problem is largely overcome through the use of embedded disk buffers used as a cache for the tape library. The largest growth area for magnetic systems is in long-term data retention and archival storage. Government regulations and numerous other legal issues are now mandating that certain types of data be retained in a machine readable format forever.

The majority of today's tape systems use linear recorded tape. Helical scan magnetic tape systems, developed and produced by Pacific Rim manufacturers and led by Sony. have made significant inroads into the tape market over the last few years and totally dominate the digital video recorders markets. Linear tape cartridges have begun to mirror the >60% annual areal density growth rate of magnetic disk systems over the last decade through two key breakthroughs. These are the incorporation of magneto-resistive heads and track following servos. Five years ago, several tape cartridges were needed to backup a single disk drive; today, a single tape cartridge can backup any disk drive. These two technologies have allowed linear tape systems to approach the areal density capabilities of helical scan magnetic tape systems. Challenging aspects of maintaining this growth relative to magnetic disk are the twin requirements of media interchangeability and backward compatibility

Optical Storage Futures

Optical (laser) recording technology has fallen far behind magnetic recording and progress remains well behind magnetic storage. Most of the optical disk activity today remains centered in the CD, DVD and video stream technologies. WORM optical storage shipments are projected to reach zero in the next two years and give way to logical encrypted-key WORM features magnetic tape technologies. It has become the ideal consumer digital content technology Capacity and performance improvements will continue to evolve, though significantly slower than magnetic storage.

Optical storage is no longer a data center technology. Current optical developments are focused on the consumer, entertainment, and content development markets. The arrival of blue-laser recording is not far off and should offer 20-30GB per cartridge. The long-standing optical storage industry challenges of format standards should again continue to slow optical adoption rates.

Holographic storage has been in development for over twenty years but hasn't become commercially available. Several companies continue to work on the technology with the media, all proprietary WORM remaining as the biggest critical path challenge and primary product differentiator. Holographic storage uses the entire thickness of the media, unlike magnetic disks that only use the surface on which to record data, creating volumetric or a 3-D recording efficiency. The most likely use of initial WORM holographic products may be as follow-on to current DVD technology as the data rate and capacity of holographic extends well beyond present DVD limits.

Optical tape products have also been under various types development for years with no commercial market yet established Capacities range from 500 gigabytes to over one terabyte per cartridge non-compressed using a standard 5.25-inch cartridge package. Optical tape projects are aiming at the WORM, scientific, streaming media and long-term archival markets. The WORM capability is now being provided as an option on magnetic tape drives. Optical storage has been the technology of the future, and it always will be.

Technologies on the Horizon

MRAM (Magnetic Random Access Memory) is a microscopic memory-cell technology that consumes little power and is non-volatile memory, meaning it retains data when power is shut off MRAM is much faster than Flash memory. The current speed projection for MRAM is about six times faster than today's DRAM memories with initial chip capacity projected to be 1 megabyte. Still under development, MRAM depends on magnetic polarity to store data, rather on electricity like DRAM chips, and stores bits in magnetic layers rather than in charges, yielding non-volatile solid-state storage with speeds comparable to SRAM. MRAM is now considered to be cheaper to manufacture than DRAM and SRAM chips. Initial production will use .25 micron technology, with limited production expected to begin in 2004 from a variety of manufacturers.

Data recording has seen few limits in the past decade. A new nano-technology called Millipede--under development at IBM--has demonstrated storing data at a density of a trillion bytes per square inch, about 20 times denser than magnetic disks available today. Data is represented by punching microscopic holes into a thin plastic film coated media. The nano-tip probe uses thermo-mechanical data writing to create a 10-nanometer (one millionth of a millimeter) hole. At about 400 degrees centigrade, the nano-tip comes into contact with the plastic substrate allowing it to "write" by punching a hole into the surface. Millipede also appears to be quite energy efficient and is aimed initially at the mobile computing market. If data transfer rates can be improved, it may become a data center technology. Production remains several years away.

Carbon nanotube transistors (CNT) are hexagonal tube-shaped molecules made of pure carbon atoms just three atoms thick. These tubes are light, yet strong, and show much better electrical and heat conductance characteristics than copper. CNTs are now being viewed as the most likely replacement for silicon chip development when silicon reaches the limits of the physical line size in ten years or more. Many questions about the manufacturability of CNTs have not been answered, but the potential for CNTs is appealing. In parallel, efforts are underway to extend the use of silicon technology as Moore's Law for semiconductor improvement continues to prevail.

Other technologies on the distant horizon include Flourescent Multilayer Discs, SOEs (Sub-wavelength Optical Elements), photonic computing, neutral networks, fuel cells as a battery replacement, and MEMs (Micro-Electromechanical Systems).
COPYRIGHT 2003 West World Productions, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Moore, Fred
Publication:Computer Technology Review
Date:Jul 1, 2003
Previous Article:Networked storage management software hits 10-year mark: thanks to IBM.
Next Article:SAN mid-year report card: the enterprise is the activity hotbed.

Related Articles
Without A Revolutionary File Management Solution There Can Be No True SAN.
Archival data has a new mission: Critical; it's not what it used to be.
Storage resource management: the next generation; elementary storage resource management graduates into business-centric, policy-based automation.
Virtualization: a strategic tool to beat storage inefficiency. (Storage Networking).
Goodbye to old, hello to the new SRM confusion: enterprise storage resource management fulfills the promise.
Extending SRM with active archiving to manage the data life cycle. (Managing Data).
No quick virtualization fixes: achieve the goals of virtualization through holistic storage management.
Information infrastructure: the new IT discipline.
Information lifecycle management: the next wave.
The network-centric file management appliance: overcoming the challenges of enterprise file services.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters