News

Fastest Xeon to Date Built for 2-Way Servers

Intel Corp. hit the 800-MHz mark with its latest Pentium III Xeon processor today. The speediest chip to date is designed for workstations and what the microprocessor giant refers to as "front-end" servers, a class of single- and dual-processor machines for functions such as Web site hosting and e-commerce authentication.

The processor, built on the 0.18-micron process technology, features a 133-MHz systems bus, and 256 K Advanced Transfer Cache memory with Advanced System Buffering to boost performance by placing a full-speed L2 cache memory directly on the processor die and increasing the width of the data pathway to the processor.

According to Pat Buddenbaum, the IA-32 marketing manager for Intel’s (www.intel.com) server computing division, features such as thermal monitoring, asset management and on-cartridge voltage management for increased system reliability were brought down from higher-end server chipsets and added to the 800-MHz processor.

But the new processor won’t show up in 4- and 8-way servers, which still sell with 550-MHz processors.

"This is really specifically designed for 2-way servers, and it won’t work or run in 4-way or 8-way systems," Buddenbaum says.

Buddenbaum cites the front-end system bus as the most significant technical hurdle causing the processor to remain in single and dual processor systems for the time being. A chipset with the 133-MHz bus, 256K and L2 cache won’t scale well past 2 processors.

"The Xeon family is becoming segmented between 2-way systems and the 4- and 8-way families," Buddenbaum says.

The 2-way machines are targeted at tasks such as caching, firewall and DNS authentication, while the higher-end servers are tuned for heartier chores such as OLTP, ERP programs, e-commerce and data warehousing.

Beyond technical difficulties, Buddenbaum says Intel had business reasons not to incorporate the latest Xeon into 4- and 8-way boxes.

"In the higher-end systems, we didn’t want to introduce a brand new platform for our OEM’s and customers to qualify. For us to do that today, it just didn’t make sense," he says. -- Thomas Sullivan

About the Author

Scott Bekker is editor in chief of Redmond Channel Partner magazine.

Featured

  • World Map Image

    Microsoft Taps Nebius in $17B AI Infrastructure Deal To Alleviate Cloud Strain

    Microsoft has signed a five-year, $17.4 billion agreement with Amsterdam-based Nebius Group to expand its AI computing capabilities through third-party GPU infrastructure.

  • Microsoft Brings Copilot AI Into Viva Engage

    Microsoft 365 Copilot in Viva Engage is now generally available, extending Copilot's AI-powered assistant capabilities deeper into the Viva platform.

  • MIT Finds Only 1 in 20 AI Investments Translate into ROI

    Despite pouring billions into generative AI technologies, 95 percent of businesses have yet to see any measurable return on investment.

  • Report: Cost, Sustainability Drive DaaS Adoption Beyond Remote Work

    Gartner's 2025 Magic Quadrant for Desktop as a Service reveals that while secure remote access remains a key driver of DaaS adoption, a growing number of deployments now focus on broader efficiency goals.