A new server vendor this week launched its first systems, the cornerstones of the companys vision of a highly flexible and dynamic data center infrastructure.
Officials with Fabric7 Systems Inc. claim that the new Q160 and Q80 systems, combined with the partitioning and management capabilities from the Mountain View, Calif., company, will help build data centers in which resources can be shared and provisioned according to application demand and business needs, without interfering with the ongoing operations. They refer to it as a “fabric computing architecture.”
/zimages/4/28571.gifIBM brings 10 Gigabit Ethernet technology to its xSeries servers.Click hereto read more.
“The whole focus is on developing a product line that allows dynamic IT,” said Sharad Mehrota, president and CEO of Fabric7.
Aimed at the midrange to high-end markets served by traditional Unix vendors such as Sun Microsystems Inc., IBM and Hewlett-Packard Co., the new systems are highly scalable servers based on Advanced Micro Devices Inc.s single- or dual-core Opteron processors and rely on hardware partitioning and the virtualization of the I/O. They also rely on AMDs Direct Connect architecture to reduce latency.
Fabric7s Q-Par technology enables users to carve up the chassis into separate servers, and provision them as needed. Those partitions work with I/O devices and storage components, and all can be managed with Fabric7s Q-Visor management software.
The Q80 allows for up to four partitions per chassis; the Q160 up to seven. Both can run as up to 16-way servers, and both run Linux from Red Hat Inc. and Novell Inc.s SUSE unit or Microsoft Corp. Windows environments.
The Q160 offers up to 128GB of memory per server and up to 30G-bps I/O capacity. Its available immediately, starting at $144,000. The Q80, with up to 64GB of memory, will be available in January, starting at $42,000.
/zimages/4/28571.gifCheck out eWEEK.coms for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.