Advanced storage roadmap - Storage Management

Computer Technology Review, Sept, 2003 by Fred Moore

A real density (or bit density), the amount of data that can be stored in one square inch of surface area of recording media, has become the most significant factor in the rapid evolution of magnetic storage technologies. Magnetic recording is clearly the dominant recording technology today and for the foreseeable future. Using magneto-resistive (MR) and Giant MR Heads (GMR), magnetic disks surpassed the areal density of optical disks in early 1995. Disk recording density in the laboratory jumped from 35.3 gigabits in 1999 and went over the 100 gigabit per-square-inch level in 2002. Long-range density projection approaches 1 terabit per-square-inch for magnetic disks.

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. Recent introduction of a 1-inch form factor may point the way to future configurations of the disk drives. There are signs that 2.5-inch form factor disk drives may start to displace 3.5-inch drives for many applications due to the increasing areal recording density, unlimited data storage requirements for many applications, as well as constraints on available space in consumer electronics devices.

Magnetic recording for both tapes and disk encountered few real limits in the 1990s as witnessed by a 60-plus percent compounded annual areal density growth rate. Within the next five years, it is anticipated that magnetic disks may encounter a physical barrier known as the super-paramagnetic limit where magnetic particles are so close together that they interfere with each other. So far, this limit has not been reached and may be avoided by using new technologies.

It is likely that the rate of data storage areal density increase in magnetic recording will start to drop gradually due to the greater difficulty in making the new technology work. A particular technical challenge will be to make 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.

Advanced Storage Architecture

Storage Networking requires successfully integrating many components that are evolving at different speeds but are all aiming at delivering the information utility class levels of service when they come together. Many of the technological barriers for storage networks, including performance and scaling, are falling. The optimal advanced storage architecture includes the fusion of large pools of storage subsystems with network-attached clients connected to heterogeneous servers. At the center of this architecture is the future version of today's network switches and directors, the Storage Domain Director. The advanced, fault-tolerant storage architecture enables centralized management of distributed computer resources finally to become a reality.

As the 'brains of the SAN,' the Storage Domain Director, evolves it provides numerous long-overdue storage functions. These include controlling SAN traffic, centralized storage management, storage consolidation, and outboard data movement between disk and tape subsystems (server-less functions). In addition, heterogeneous data sharing of a single unique instance of data, and non-disruptive scalability and configuration, enables the long-awaited capability for proactive and people-transparent storage management. The ability to perform block and file operations simultaneously and thus bridge the 'number and name' worlds within the same storage subsystem is also a most strategic and desired outcome of this advanced architecture. Emerging concepts such as IP storage, advanced SRM, and in-band (symmetric) or out-of-band (asymmetric) virtualization appliances are providing valuable building blocks for advanced storage architectures. Many companies are delivering pieces of the complete information utility, but the complete vision path will take several years to arrive. Nonetheless, the path to the future for storage subsystems can be described using the components that are just now becoming visible.

www.horison.com

Reprinted from Storage Manifesto with permission from Fred Moore, president, Horison Information Strategies.