Cisco Reveals New UCS Servers Featuring Intel Xeon E5 Chips

Cisco updates networking and management capabilities of its Unified Computing System

Cisco Systems has revealed a significant upgrade of its Unified Computing System (UCS) data centre platform.

Cisco first introduced UCS in March 2009, a move that surprised its rivals at the time as it signalled Cisco’s intention to expand beyond its networking heritage and into the world of data centre servers and hardware.

Third Gen Upgrade

Now Cisco has revealed that it has upgraded the performance, as well as the networking and management capabilities of UCS, to assist customers in supporting virtualisation, building clouds and deploying applications in a more timely manner.

The third generation of the UCS platform makes use of new energy efficient Intel Xeon processor E5-2600 product family, which Intel introduced earlier this week. Other performance tweaks include up to eight times the memory capacity and four times the I/O compared to previous UCS servers.

Meanwhile, the Cisco UCS Manager now allows IT administrators to manage both blade and rack servers as a common entity. In addition, the management domain has been expanded to span thousands of servers across data centres around the world (although this is only arriving in the second half of 2012).

Cisco said that UCS customers report dramatic operational and cost improvements, with some apparently claiming up to 30 percent lower infrastructure expenses; 90 percent reduction in deployment times; 40 percent improvements in application performance; and 60 percent reductions in power and cooling costs.

Fabric Improvements

In an effort to cater for these increasingly virtualised and distributed workloads, the network has also had to be improved. Cisco revealed that it has rolled out the chassis I/O module 2204XP, which offers 80Gbps and 160Gbps down to each chassis to handle workload bursts. The module also offers load balancing across all ports.

Another enhancement is the Cisco UCS 6296UP Fabric Interconnect, which doubles the switching capacity of the UCS fabric from 960Gbps to 1.92Tbps and reduces end-to-end latency by 40 percent to deliver industry-leading application performance. Cisco said that the fabric interconnect provides infrastructure agility at scale with unified ports and greater energy efficiency, lowering watts per port by 36 percent.

“Our customers care most about time-to-deploy business applications, application performance and unified management of bare metal, virtualised and private/public cloud environments,” said Soni Jiandani, senior vice president, Data Center Group, Cisco.

“Cisco pioneered fabric computing and service profiles – still not duplicated by any other vendor- which allow IT managers to deploy applications in minutes instead of days,” said Jiandani. “Now with innovations in UCS Manager their integrated management spans rack and blade servers, and scales across data centres to thousands of servers globally.”

Amid the raft of upgrades and tweaks, Cisco lost little opportunity to boast about how well the UCS platform is doing, despite initial doubts. Cisco said that since it was introduced in 2009, nearly 11,000 customers worldwide have deployed Cisco UCS to unify their data centres. It also said that UCS is the fastest growing product in Cisco’s history, with a $1.3 billion (£822m) annual run rate.