Home News Will Intel’s new CPU offer cloud service providers the processing power they...

Will Intel’s new CPU offer cloud service providers the processing power they have been waiting for?

archives2013

By David Howell, Tech Journalist and Commentator

With a quoted 45% increase in efficiency and up to 50% improvement in performance using up to 12 cores compared to the previous generation, Intel’s Xeon Processor E5-2600 v2 (Ivy Bridge-EP) product family looks set to offer what could be a quantum leap forward in CPU power for data centres looking to leverage their services for the cloud and big data applications.

Diane Bryant, senior vice president and general manager of the Data centre and Connected Systems Group at Intel commented:

“More than ever, organizations are looking to information technology to transform their businesses. Offering new cloud-based services requires an infrastructure that is versatile enough to support the diverse workloads and is flexible enough to respond to changes in resource demand across servers, storage and network.”

Cloud Power Boost?

As the cloud has rapidly evolved, data centres have struggled to offer the flexible on-demand services that business now need to power their enterprises. Partnering with IBM and their NeXtScale System, the new Intel CPU’s can now be built into dynamic platforms that can rapidly evolve as service demand changes. The new chip will also be used in the new x3650 M4 HD high-density storage server from IBM to support big data storage and analytics.

And of course with virtualisation becoming ubiquitous, Intel’s new chip has been designed with this in mind. IDC’s Digital Universe Study from last year stated that 82% of global telecommunications providers are expected to evaluate SDN (Software Defined Networking) and NFV (Network Function Virtualization) this year. Intel explained their approach:


“Using Intel’s Open Network Platform (ONP) server reference design, customers can use high-volume Xeon-based servers combined with industry open standards to consolidate virtualized networking applications. This allows them to deliver market leading throughput performance and latency for SDN and NFV workloads. Intel’s ONP server reference design is based on the Wind River Open Virtualization Profile and the Intel Data Plane Development Kit Accelerated Open vSwitch.”

For data centres and their customers the new Intel chip is the first of what has been dubbed SoC (System on a Chip) with Diane Bryant succinctly commenting at the last Intel Datacentre Day: “Today, we look at IT as the service. IT is no longer supporting the business, rather IT is the business.”

And this is at the heart of the new CPU’s that Intel has developed. The new Xeon is scheduled for release next year, but it’s interesting to see how Intel is also positioning their other low-powered chips. These are in the shape of the Atom for data centre usage. With the Atom’s fabrication process converging with the Xeon range, the 14nm Atom architecture code named ‘Denverton’ will roll out in the near future. The two streams will offer data centres the options they need to build service platforms that are affordable.

What is clear is that server architecture that couples CPU with memory for big data and cloud applications need the level of processing power that the new Xeon chips offer. Data centres are looking to differentiate in the marketplace, and know that their customers are struggling with their current hardware set up. And with virtualisation marching on to dominance, the E5 can’t come quickly enough.