Realize ROI faster with green datacenters
2^nd August 2008, PC Quest

Currently, 2% of all energy utilized in the world is thanks to datacenters. This is equivalent to the total energy consumed by the airline industry globally. Server capacity on the other hand, is expected to grow 6 times of what it was in 2000 by 2010. Consequently, storage capacities are set to grow 69 times to accommodate the growth in servers. This directly relates, despite all advances in the data centre industry, to a doubling of energy utilization every five years, and for the records, energy costs have gone up by 10% last year in the US. What is more shocking is the fact that the technology density of datacenters is growing at 20 times every 10 years, which means you should ideally have a datacenter which perform 20 times as well as your current datacenter in the next 10 years. This almost impossible proposition is worsened by the fact that 78% of all datacenters globally have not been upgraded, scaled up or enhanced ever since they were deployed, and are more than 7 years old. The only way out is to either ensure that your next data centre is green or change the business dynamics to suit the changing trend in datacenter deployment, which lowers the lifetime significantly and even provides datacenters on demand.

A green roadmap

As an ideal start, IBM announced in India recently, the launch of Project Big Green 2.0, its global initiative to help enterprises build energy and power efficient technology infrastructures that can meet growing business requirements while realizing rapid financial benefits from such investments. Last year, during phase 1 of Project Big Green, IBM invested USD 1 billion to promote the concept of energy efficiency. This year, the project takes off from the point that a higher percentage of IT businesses are seeing power and cooling as the biggest challenges in datacenter mgmt (22% more than 17% last year).

Besides server optimization and integration services for VMware server virtualization, an updated SAN volume controller and IBM Tivoli enhancements, the project sees the launch of a new concept in datacenters - on demand and personalized datacenter in a box. Aimed at large format events, that need a sudden spurt of computing power, IBM's Portable Modular Datacenter saves space, energy and is self sufficient for its running power. In other words, it provides the complete physical infrastructure including power and cooling systems and remote monitoring. The datacenter can be shipped and deployed into any environment and can support multiple technology vendors. Ideal for Indian markets that are devoid of real estate, these boxes claim to brave extreme climatic conditions, making it ideal to be deployed at remote locations as well.

Scaling up enterprise models

If your business requirements are such that it requires higher turnkeys over an extended period of time with regular upgrading, Scalable Modular Datacenters are ideal for you, with 500-2,500 sq ft, at a minimum of 15% improved energy efficiency, 20% lower price points and a deployment time of 8 to 10 weeks. This is a pay-as-you-grow model and adding additional stacks does not require too much time of downtime of existing systems. The Enterprise Modular Datacenter, on the other hand, promises an attractive energy efficiency of 66%, and an open architecture which includes multi-vendor support and service. This is ideal for time intensive and high availability operations.

An online green meeting room In order to collaboratively walk the path of energy efficiency in datacenters, a global consortium called The Green Grid was created last year, with members ranging from Dell to Microsoft to Cisco, Sun Microsystems, and Wipro Technologies being one of the most recent additions. The central aim of the consortium is to 'advancing energy efficiency in data centers and business computing ecosystems'.

The Green Grid is focused on defining meaningful, user-centric models and metrics, developing standards, measurement methods, processes and new technologies to improve data center performance against the defined metrics, and promoting the adoption of energy efficient standards, processes, measurements and technologies. The Green Grid has recently embarked on a mission to qualitatively analyze and collate the various green initiatives directed towards more efficient and eco-friendly datacenter design and management, across its 20 odd partner members.

Futuristic cooling blueprints Scientists, in collaboration with IBM have also developed a method to cool computer chips that have circuits and components stacked on top of each other with tiny rivers of water, an advance that promises to significantly reduce energy consumed by data centers. A few weeks back, Researchers, in collaboration with the Fraunhofer Institute in Berlin, demonstrated a prototype that integrates the cooling system into the three dimensional chips by piping water directly between each layer in the stack. These so-called 3D chip stacks -- which take chips and memory devices that traditionally sit side-by-side on a silicon wafer and stacks them together on top of one another-presents one of the most path-breaking approaches to enhancing chip performance beyond its predicted limits, while simultaneously reducing the energy consumed by data centers.

Seeing business sense in green An electrical engineering company suffers loss, and decides to go green. It has seen benefits at various levels-from changing the air conditioner to deploying a pay-as-you-expand datacenter architecture Bharat Bijlee is a pioneer in the electrical engineering in the private sector, and claims to be a multi-product, multi-division organization, operating in two business segments, Industrial Products which comprises transformers, motors and drives; and Contracting, ie Projects. Unlike its non-private counterparts, Bharat Bijli has embraced IT quite early in its growth. The company confesses to have learnt the importance of going green from its first infrastructure, which was housed in a server room and was inefficient in handling the expansion plans of the company. Low capacity UPS, obsolete switches, water seepage and rodents were some issues the company had to contend with. Archaic infrastructure resulted in frequent unavailability of systems & impeded productivity.

As the company grew, it has the challenge to have in place a state-of-the-art infrastructure that adheres to these requirements and meet the current as well as the future demands of the company. This led to the deployment of a secure integrated environment with high performance and improved connectivity that allows quick deployment of new generation applications and updates connected to it. The first step towards this was upgrading the existing infrastructure by taking the upgrade from a 10 mbps LAN to 100 & 1000 mbps in some cases to ensure higher availability. All connected factors like power, cooling, fire protection, smoke detection, interoperability with vendor products were implemented with heavy emphasis on issues like power breakdown. The new infrastructure housed a rodent repellent, water detection system, and precision AC as opposed to comfort AC installed earlier. The space relieved the company from the clutter of the server room. The implementation and execution of this project now allow the company to shift its focus from administration of the IT infrastructure to other value added projects more critical to them.

Bharat Bijli in association with Netsol, had quite a few challenges in seeing the project through. To begin with, the project commenced during monsoons and the site was set up at a rocky terrain. Thus, digging and cabling was a cumbersome task. As a pro-active measure, the company tested equipment during the up gradation process. This allowed detection of faults and avoid delays. At a later stage, it incorporated a scalable datacenter, foreseeing growth of certain regularity. Bharat Bijli's experience with a scaleable datacenter is such that it eliminates delays, dependency on time lines and business churn outs, which according to the senior management is the biggest reason for failure of green IT projects. A scaleable model lets corporations?big, small, or medium-pay as they grow and procure hardware that can sync up with existing systems and goes a long way in saving space, power and time. Scaleable datacenters by definition are aimed at delivering necessary power, cooling, security and monitoring capabilities for your data center. These required elements can ensure efficient & reliable server operations & a security-rich environment- ensuring hardware is in an optimal setting all time, reducing power leakage to a great extent.