eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
Intel Corp.s decision to compete head-to-head with Advanced Micro Devices Inc. by bringing 64-bit capabilities to its Xeon processors should result in a greater number of hardware and software options for enterprises looking for a gradual transition from 32-bit to 64-bit computing.
Craig Barrett, CEO of the Santa Clara, Calif., company, last week at the Intel Developer Forum announced plans to offer 64-bit extensions in its 32-bit Xeons. It will start next quarter with a 3.6GHz Xeon, code-named Nocona, and early next year, a multiprocessor Xeon, code-named Potomac, will offer 64-bit capabilities.
The extensions will enable systems with Xeons, like those already equipped with AMDs Opteron processor, to run 32-bit or 64-bit software. Both use Intels standard x86 architecture.
Industry observers expect that ISVs will be less reluctant to create software targeted for 64-bit systems built on x86 now that the worlds top chip maker is in the market. Indeed, Microsoft Corp., of Redmond, Wash., announced last week that Windows will support such systems in the second half of the year and that Linux vendors Red Hat Inc. and Novell Inc.s SuSE Linux unit will support them as well.
“Its a win for customers, and its a win for people who have embraced the Intel architecture,” said Joshua Levine, chief technology officer and administrative officer for E-Trade Financial Corp. and a critic of Intels 64-bit Itanium processor, which doesnt use x86. “Im hoping this is really now something that Intel will see as permanent as people want … upward compatibility forever.”
Levine, whose data center includes Opteron-powered servers, said the appeal of Intel is that systems are compatible through the 32-bit space—and now 64-bit computing.
“You can run your applications on any machine,” said Levine in New York. “Thats the power of compatibility. You dont have to port; you dont have to provision. If you have a lot of machines in the data center, what an awful thing it is to have to say, We have to provision this machine and … this machine.”
AMD, of Sunnyvale, Calif., said there are still key differences between Opteron and the extended Xeons. Key among those are the memory controller, which is built directly onto the Opteron chip, and the chips HyperTransport technology, which speeds communications among processors. AMD also is further along in developing its Opterons. Last week it unveiled two new low-power versions for thin servers.