HPCwire: SeaMicro Launches Low-Power Xeon Server

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

January 31, 2012

SeaMicro Launches Low-Power Xeon Server

Increasing server bandwidth while slashing power and space across the scale out data center

SAN FRANCISCO, Calif., Jan. 31 -- SeaMicro, the pioneer in energy efficient servers, today announced its foray into the world of brawny compute with the immediate availability of the industry's first fabric-based Intel Xeon micro server, the SM10000-XE. SeaMicro has once again defined best in class by building a micro server that uses half the power and one-third the space, but for the first time is targeted at heavy weight workloads as opposed to the light weight ones that are traditionally associated with micro servers. By offering both Intel Atom and Intel Xeon based micro servers, SeaMicro is the only server company able to deliver low power high density solutions to all parts of the scale out data center.

"SeaMicro has invented technology that allows our SM10000 family to address all segments of the scale out data center," said Andrew Feldman, CEO of SeaMicro. "The addition of the SM10000-XE to the SM10000 family allows one architecture to support big and small processors -- enabling customers to match their workload to the processor best suited for that work. The SeaMicro SM10000 family delivers a single architecture capable of supporting different processor types, each optimized for a different type of work."

The SM10000-XE is built around SeaMicro's new Freedom Fabric ASIC -- the industry's first and only second generation fabric chip. The Freedom Fabric ASIC contains three key patented technologies:

SeaMicro's Input/Output Virtualization Technology eliminates all but three components from SeaMicro's motherboard -- CPU, DRAM, and the ASIC itself -- thereby shrinking the motherboard and reducing power, cost and space.
SeaMicro's new TIO (Turn It Off) technology enables SeaMicro to further power-optimize the motherboard by consolidating functionality and turning off unneeded CPU and chipset functions. Together, SeaMicro's I/O Virtualization Technology and TIO technology produce the smallest, most power efficient motherboards available.
The Freedom Supercompute Fabric ties these small, power reduced motherboards together with an industry leading 1.28 terabits per-second of bandwidth. The Freedom Supercompute Fabric is built of multiple Freedom ASICs working together to create a multi-dimensional torus -- delivering a low latency, massive bandwidth fabric with exceptionally low power draw.

"SeaMicro is the only company that can bring the power and space benefits of the micro server to all segments of the scale out data center," said Zeus Kerravala, senior vice president and distinguished research fellow at the Yankee Group. "The Freedom Supercompute Fabric is the only fabric that can support large and small workloads, enabling SeaMicro to plot a trajectory of servers that leaves no segment of the data center untouched. Their rate of innovation is simply dazzling."

The SeaMicro SM10000-XE

The SM10000-XE contains 64 quad-core Intel Xeon processor E3-1260L CPUs for 256 2.4 GHz cores in a 10 rack unit system for 1,024 cores in a standard rack. The SM10000-XE delivers 10 gigabits of bandwidth to each quad core processor -- setting the high watermark for bandwidth per unit compute.

A single SM10000-XE replaces 32 dual socket servers, but draws half the power and takes one-third the space -- all without requiring any changes to software. Like all SeaMicro servers, the SM10000-XE is exceptionally easy to use. It simplifies data center operations and management by eliminating layers of Ethernet switches, server management devices and expensive load-balancers. It is plug and play -- customers can deploy it without modifications to existing operating systems, application software or management tools.

SeaMicro's SM10000-XE system is comprised of:

64 x86 quad-core 2.4 GHz Intel Xeon processors
256 cores per 10 rack units, 1,024 cores per rack
10 gigabits of bandwidth to each quad core processor
Up to 2.048 terabytes of Samsung's Green DDR3 DRAM -- 32 GB per processor
Up to 64 SATA solid state or hard disk drives
Up 16 10 gigabit Ethernet links or up to 64 one gigabit Ethernet uplinks.

Additional benefits include:

support for 64-bit operating systems
industry leading bandwidth per unit compute
runs off-the-shelf OSs including Windows
simplifies data center infrastructure by eliminating top of rack switching, load balancing and console servers.

The Processor and Memory

The SM10000-XE uses the most efficient quad-core Intel Xeon processor E3-1260L and the most advanced, power-efficient DRAM -- Samsung's 1.35V, 30nm class, Green DDR3. The Intel Xeon processor E3-1260L has a 45W TDP envelope, 2.4GHz CPU core based on the Sandy Bridge microarchitecture, features 4 cores, supports 8 threads with Intel Hyper-Threading Technology, and is able to increase its clock speed up to 3.3GHz with Intel Turbo Boost Technology 2.0. It provides 30 percent better performance per watt compared with prior generation processors.

Samsung's high density Green DDR3 achieves more than a 70 percent power savings over 1Gb 1.5V 50nm class DDR3, contributing much to an improved TCO. Further, the DIMM's unusually small form factor (67.6x30mm) provides additional space saving benefits from a board real estate standpoint, therein increasing system design flexibility.

Availability

The SeaMicro SM10000-XE is now generally available in the U.S. and at select international locations. The U.S. list price for a base configuration is $138,000. More information on the SeaMicro and its revolutionary SM10000 family of servers can be found at http://www.seamicro.com.

About SeaMicro

SeaMicro is transforming the data-center landscape by building servers that draw a fraction of the power and space of traditional servers. By delivering breakthrough innovations born of multiple technology domains -- CPU design, virtualization, supercomputing, and networking -- SeaMicro has created a new server category, called micro servers, specifically designed for scale-out infrastructures such as those found in the Web tier, online gaming, searching, and index computation. SeaMicro was founded in 2007 and is backed by Khosla Ventures, Draper Fisher Jurvetson, Crosslink Capital, and leading public corporations. The company also received the largest Department of Energy grant given to a server company, $9.3 million, as part of the American Recovery and Reinvestment Act.

-----

Source: SeaMicro

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...