Business Cloud News
Intel said the Xeon E7 v2 family will nearly triple performance for in-memory computing and big data applications

Intel said the Xeon E7 v2 family will nearly triple performance for in-memory computing and big data applications

In a bid to capitalise on the apparent stampede towards Big Data in all walks of life, Intel unveiled an updated Xeon chip architecture this week that will, according to the chip maker, double performance over the previous generation of server chips.

The company said the E7 v2 family, designed to support standard 32-socket servers and packing up to 15 cores with support for up to 1.5TB of memory, is being pitched at service providers and enterprises looking for more power for big data applications and in-memory computing (retail, healthcare, banking and the like); it packs triple the memory capacity of the previous generation.

“Organizations that leverage data to accelerate business insights will have a tremendous edge in this economy,” said Diane Bryant, senior vice president and general manager of Intel’s datacentre group.

“The advanced performance, memory capacity and reliability of the Intel Xeon processor E7 v2 family enable IT organizations to deliver real-time analysis of large data sets to spot and capitalize on trends, create new services and deliver business efficiency,” she said, adding that the technology helps reduce bottlenecks in data processing.

According to Intel the big data technology and services market is expected to grow 27 per cent annually through 2017 to reach $32.4bn and the chipmaker wants to remain at the core of the servers powering the revenue-generating services built on top of all that data, which will only grow as the Internet of Things continues to take off.

The company has already inked deals to ship the new chips to a range of vendors and cloud service providers, and said companies like eBay are already trialling the new chip architecture using the popular in-memory platform SAP HANA with and yielding greater performance (for eBay specifically, memory scoring 110,061 queries per hour on its existing baseline vs. 218,406 queries per hour with the new chips).

Beyond big data and the Internet of Things, Intel is looking to ensure its relevance in the cloud era by convincing cloud service providers to tout the “added performance benefits” of using Intel-based instances over, say, rival AMD, which has a relatively low share of the server market. Nevertheless, last month’s deal, a marketing coup according to some, ensured some equivalent of that shiny “Intel Inside” sticker branded on countless PCs will live on in the utility computing era.

@BizCloud
News