Unified Computing Overview

Transcript

Introduction

David Lawler: Hello. I'm David Lawler, VP of platform product marketing at Cisco, and I'd like to give you an overview of Cisco's vision of Unified Computing.

Interview

Today's enterprise data centers need to be more responsive than ever to changing business dynamics. Organizations must be flexible and agile to stay competitive in this challenging economy, and at the same time, they need to watch every penny and invest only in solutions that deliver ROI (return on investment) quickly.

The Cisco Unified Computing System, or UCS, marks an evolutionary next step in the realization of the Cisco Data Center 3.0 vision. The Unified Computing System provides a new architectural paradigm that incorporates advanced server technology, next-generation networking, virtualization resources and integrated management. It leverages the cutting-edge technologies of virtualization and unified fabric to combine computing, network and storage resources into one seamless system.

Unifying the data center reduces total cost of ownership and increases application performance while providing IT with an unprecedented amount of flexibility.

Why unified? Right now, data centers are full of separate systems -- servers for network for computing, storage systems for archiving and data retrieval, and networking switches for connecting everything together. Companies are starting to explore virtualization strategies, especially for servers, to help reign in expenses and keep power consumption and air-conditioning costs under control and move towards automation.

The problem is virtualizing servers only does part of the job and introduces management difficulties as well. The Cisco UCS applies the same concept -- abstract the function from the underlying hardware -- and extends it across the data center. The Cisco UCS outperforms and out-scales traditional data center architectures, and it allows companies to provision and deploy resources more quickly and more efficiently than ever.

Let's take a look at the components that make up Cisco UCS. Cisco UCS B Series blade servers are based on Intel Xeon 5500 Series processors. They can run enterprise applications reliably, adapt smoothly to changing workloads, scale energy consumption intelligently and deliver best-in-class virtualization. These blades are designed for compatibility, performance, energy efficiency, manageability and unified I/O (in/out) connectivity.

The Cisco UCS 5108 blade server enclosure houses the UCS B Series blade servers and up to two 2100 Series fabric extenders. The simple 6RU enclosure supports up to eight half-slot or four full-slot blade servers with four power supplies and eight cooling fans.

The Cisco UCS 6100 Series fabric interconnects are a family of line-rate, low-latency, lossless 10-Gigabit Ethernet, Cisco Data Center Internet and fiber channel-over-Ethernet interconnect switches that consolidate I/O at the system level.

The Cisco UCS 6120 XP 20-port fabric interconnect is a 1RU fabric interconnect featuring 20 fixed ports that can be provisioned as uplink or downlink ports.

The Cisco UCS 6140 XP 40-port fabric interconnect is a 2RU fabric interconnect featuring 40 fixed ports that can be provisioned as uplink or downlink ports.

The Cisco UCS 2100 Series fabric extenders are found in the back of the enclosure, and they extend that I/O fabric into the blade server enclosure, providing up to four 10-Gigabit-per-second
connections between the blade servers and the fabric interconnects, simplifying diagnostics, cabling and management.

And, finally, the Cisco UCS manager integrates all of these pieces into a seamless whole. This powerful Cisco UCS management software can manage up to 320 blade servers as a single system using the intuitive GUI (graphical user interface) with both command-line and XML API options, enabling rapid provisioning and configuration of resources.

With Unified Computing, your company will reap certain benefits such as business agility with features such as an intuitive standard-based management that enables customers to easily provision services in minutes versus hours or days.

Next, reduced cost of ownership. There's one-third fewer components, such as switches, adapters, management modules, servers, as well as reduced points of management, so we deliver both CapEx (capital expenses) savings, and more importantly, significantly lower OpEx (operating expenses).
Also, the system's highly scalable, with the ability to scale up to 320 discrete servers and give applications access up to 384 Gigabytes of memory in a single server.

And finally, seamless integration, since the UCS unifies compute, network, management and virtualization resources into a seamless single system.
In summary, Cisco Unified Computing System provides the integrated framework that today's enterprise data centers need to boost agility and control cost.

For more information, please visit Cisco.com/Go/Unified Computing. Thank you for watching.