Un-Weave the Hype Around Fabric Computing
Bernie Trudel, Cloud CTO, Cisco Asia Pacific
The tech sector loves a new buzz word. Company slogans or hyped-up products also come and go. I believe it's good to stay sceptical of what gets thrown around because when a truly powerful shift happens, you can see it more clearly. We are dead-centre in one of the greatest recent shifts in tech-infrastructure in recent times: unified data centre fabrics.
Imagine you wanted to get a coffee in the morning but to get the coffee you had to cross the street to buy it. Once you got there and paid, you had to go over the street to get your change, leaving your coffee at the shop. By the time you got back it was cold and they had to make you another. Instead imagine you pulled out your smartphone, opened a coffee-app, selected your order, the app automatically charged you; then your coffee was delivered quickly and hot. This is the shift to unified data centre fabrics.
Gartner has reported that fabric computing is on the radar for several IT groups and we are seeing companies across the region migrating to cloud based operations built on fabric data centres. (Gartner report: Fabric Computing Poised as a Preferred Infrastructure for Virtualization and Cloud Computing, February 11, 2011, George J. Weiss and Andrew Butler Report. ID number: G00210438.)
It's worth taking a step back and finding out what the fabric in unified data centre fabrics is, to grasp why this is such a significant trend. Fabric computing weaves together compute, networking, storage access and virtualisation into one cohesive system. The term 'fabric' comes from its appearance; it is a visual weave of parts and cables.
The reason why fabric computing is so powerful is because it brings traditionally disparate, siloed processes into one unified system. Fabric allows you to reconfigure all system components at the same time. This reduces management time and cost by allowing you to manage previously disparate systems, holistically.
Previously, data centres supporting cloud systems were large, complex and had numerous physical parts. This meant cloud data centres took up large amounts of space and required large amounts of power to run. A very expensive set up, difficult to upgrade. Fabrics are optimised for virtualisation; this means that whenever you add new blade or rack server to your data centre, the time it takes for the server to come online is reduced from hours to minutes.
Fabric computing consolidates data centres by reducing what you physically have on premise and then connecting what you have with high-bandwidth connections. This saves both space and cost by increased energy efficiency and also allows servers to do what they do best, deliver computing power.
Let me go through two examples. Let's start with KPIT Cummins, an end-to-end product design, engineering and IT services company in India. KPIT Cummins was looking to improve the utilization of IT assets and physical infrastructure and provide flexibility for users to work from anywhere without compromising security. One way to address the problem was to detach physical infrastructure from data, applications and compute through 'virtualization.' KPIT decided to implement virtualization at the desktop level. The datacentre architecture and the UCS framework from Cisco along with the VCE Vblock platform appeared to be ideal.
KPIT deployed the Vblock platform for a virtual desktop interface (VDI) implementation to serve 1200 users across India. The Vblock 1, designed for large virtual machines was deployed in a compact footprint and provided for a mid-sized configuration to deliver a broad range of IT capabilities.
After conducting a POC, KPIT discovered that the solution provided an interconnect between storage and networking along with good levels service support. The solution enabled scalabilty, interoperability and ensured a single point of contact for support, to the customer.
Post implementation, KPIT managed to obtain optimum performance and right size its infrastructure in tandem with changing business needs. The solution design enabled KPIT to move a number of core applications on to the VDI platform in a short period of time. In addition to a transition to private cloud infrastructure, the Vblock platform offered KPIT processing, network performance and storage capacities which support enhanced security and business continuity.
The solution helped KPIT manage the growing deluge of smart devices, data and applications along with an increasingly mobile workforce and constrained IT budgets. VDI deployment has also helped enhance user experience and improve employee productivity by enabling end-users connect with business critical data and applications via any device, on demand .
Since the servers are designed for the cloud, KPIT can add on virtual desktops with ease while also reducing costs, carbon footprint and administrative personnel. The solution is scalable, flexible and helps achieve operational efficiency with growth, reduces downtime and improves business agility. Importantly VDI has enabled central management of data and applications furthering the security levels of access.
Consider another example of Geometric, a Global Engineering services and Digital Technology solutions company from Mumbai. Geometric was grappling with issues pertaining to conventional desktop deployment. Individual personalization done on PCs prevented standardization, increased support requirements and software related errors. This affected security; manual scrutiny was needed to ensure data protection. That was time-consuming and caused significant downtime.
To overcome these issues, Geometric decided to implement desktop virtualization and chose a Cisco solution based on the Virtualization Experience Infrastructure (VXI) architecture. The solution was installed at the Geometric data center at Pune for a POC. The POC enabled Geometric to validate the performance of the solution, do a better sizing and demonstrate the same to their applications teams.
The solution enabled Geometric to understand first hand, the benefits of a virtualized infrastructure. It helped reduce the overall cost of deployment besides reduction in IT, power and cooling costs. Since virtual machines can be standardized and cloned effectively in a short time, Geometric could reduce deployment time for new requirements and IT support efforts.
Geometric also managed to reduce the number of networking components and their complexity-signified by a reduction in the number of cables to be managed. The solution platform could be easily integrated and managed with Geometric's existing VLAN structure; modular investments helped to add components and step up the performance gradually rather than through massive initial investment. After completing the initial deployment of 250 desktops, Geometric obtained significant benefits in TCO which will improve with scale.
Clearly the unified fabric isn't just a buzz word or hyped-up product. It is the next step to transform IT departments into active business profit centres, by allowing infrastructure and applications to be deployed rapidly and efficiently. A collaborative solution at its best!