ARN

Oracle takes apparent step toward standardizing on a single chip design

Analysts say the unveiling of midrange and high-end servers running Oracle-built Sparc chips that share the same architecture is a step toward fulfilling the company's post-Sun plans.

Oracle's Unveiling of a batch of servers based on new Sparc processors late last month marked what could be the start of an expected shift toward standardizing the vendor's two families of Unix servers on a single chip architecture.

Oracle currently sells two lines of Unix servers: the midrange T-Series, based on Sparc processors designed in-house, and Fujitsu-built high-end M-Series machines running the Japanese manufacturer's Sparc64 chips.

On March 26, the same day it introduced a batch of new T-Series boxes, Oracle unveiled the first M-Series server that it had designed -- and the first to run Oracle-built Sparc processors. "This is all Oracle [intellectual property]," said Marshall Choy, the vendor's director of systems solutions and business planning.

Observers have been expecting Oracle to move the two lines to a single chip architecture since it finalized its acquisition of Sun Microsystems in early 2010. Analysts have said that such a move would reduce hardware and software development costs at a time when Oracle hardware sales are declining.

"This is really the first instance of them delivering on this promise," said Nathan Brookwood, an analyst at Insight64.

Choy would not confirm that the launch of the new M-Series server marks the first step of an Oracle plan to have its own Sparc chips run all or most of its symmetric multiprocessor (SMP) systems, though analysts contend it's been the vendor's strategy for a long time. "They've been saying they would do this for years," Brookwood said.

The new high-end M-Series SMP server runs a new six-core Oracle processor called the Sparc M5, and supports up to 32TB of system memory. The Sparc M5 chip shares the same core as the new 16-core Sparc T5 processor that runs the five new T-Series servers.

The difference between the T5 and M5 processors is that Oracle has removed some of the cores on the M5, which aren't as useful in SMP systems. Engineers also added much bigger Level 2 cache memory -- six times bigger than the T5, Choy said.

Brookwood said Oracle's single architecture strategy is comparable to the one Intel pursues with its x86 cores. Intel generally develops one processor core and adjusts the number of cores per chip, the cache size and other features for specific servers, he said.

A single chip architecture also fits Oracle's preference for designing entire bundled systems -- chips, operating systems, databases and applications -- in-house. The company argues it can build higher-performance systems that way, though some observers point out that such an approach ties customers to an all-Oracle stack of hardware and software.

"It's the right strategy for them; what they're really selling is integration," said Joe Clabby, an analyst at Clabby Analytics.

But he also wondered whether developing microprocessors is still a viable pursuit for Oracle. " They just came off a bad quarter; how much longer can they keep investing in this?" Clabby asked.

Brookwood, however, said moving to a single chip design could save money and free up funds for investment in Sparc chip development. "Don't forget that Oracle sells a lot of software wrapped around these systems," he said.

Niccolai is a reporter with the IDG News Service.

This version of this story was originally published in Computerworld's print edition. It was adapted from an article that appeared earlier on Computerworld.com.

Read more about hardware in Computerworld's Hardware Topic Center.