Why 64-bit computing is faster

Computers that use Intel Corp.'s new Itanium microprocessor will be much

faster than current PC servers for several reasons, though one of them is

not the common misconception that 64-bit chips are twice as fast as 32-bit

chips.

The most significant benefit of 64-bit computing is the amount of addressable

memory that the architecture supports. Current 32-bit computers are limited

to 4G of addressable memory, whereas 64-bit computers support virtually

unlimited amounts of addressable space, said John Tescatore, Dell Computer

Corp.'s director of architecture for enterprise servers.

With 32-bit systems, processing jobs that involve large data structures

must be broken into smaller, more manageable pieces, which takes more time.

"Many applications can run faster if they can put more of the application

in memory, not on hard disk," said Mike Fister, vice president and general

manager of Intel's enterprise platform group.

Another important, more subtle performance-boosting factor is how the

new chip handles the instructions that flow through it as it processes data.

The Itanium chip can process multiple instructions in parallel, reducing

the time it takes to complete jobs. This so-called Explicitly Parallel

Instruction Computing architecture was created by Intel and Hewlett-Packard

Co.

Itanium is the first product in what will be Intel's IA-64 family of

64-bit chips. The second product in the family, code-named McKinley, is

expected to ship in the second half of next year and will reportedly offer

twice the performance of Itanium.