The year was 2008 - we at Hero Honda Motors Ltd. the world’s largest two wheeler manufacturer, had been witnessing phenomenal growth. The IT enterprise was working closely with business to put in place technology-led solutions throughout the value chain. We at IT were looking at our plans and working out areas where IT would help the organisation sustain this growth and help the organisation stay ahead of its competition.

When we reviewed our plans to analyse the infrastructure requirements in terms of servers and peripheral equipment required for the next 6-8 months, we were startled to see our infrastructure and consequent data centre space, energy requirements and the related needs. It was even more worrying because we had just set up the new data centre and it was very evident that soon we would run out of space in the new data centre and would also have to enhance our power and cooling requirements significantly. All the capacity planning computations which had been done by us only a year back when the market scenario was quite different, now looked unrealistic and this made us think about ways to deal with the situation and strategise to curb our increasing infrastructure and operational costs.

One of our guiding philosophies has been that technological advancements should always attempt to make life simpler and efficient and we thought to ourselves that when we can use technology to do that for users why not attempt to achieve the same goals for the IT team as well to combat the challenge at hand.

The team was given a simple task, which was to think of the best possible way of coming out of this situation. Our teams started evaluating various options – both technological and non-technological – including ideas like consolidating applications, hosting some of the applications with third parties and looking at virtualisation.

Virtualisation was the most talked about technology and it seemed like the most fitting solution for us. Virtualisation as a concept was not new to us. For our Unix based systems, we had been using the native features of the RISC (Reduced Instruction Set Computing) based systems for virtualisation and were quite familiar with partitioning and other ideas. But now we had to explore its use for a different area which was our Windows-based applications. We started with seeing whether the learnings and experience which we had of using the technology in one set of our environment could help us in deploying a similar concept in the other set as well.

In the attempt to understand and then make our decision we did some research on the technology, we read on the Web as to what people were saying about virtualisation, we gathered case studies of any implementation so as to learn from them, we also read analyst reports and went through vendor presentations. What we read gave us more reasons to not take the plunge than to go ahead. The extent of server virtualisation globally was very low, storage and others was even lower. Article on article about numerous challenges kept staring at us; there were articles on concerns around the management complexity after the deployment; some were of the opinion that the nature and number of application instances and quite frequently OS instances remain virtually unchanged or may even go up in most of the organisations. Some mentioned that in fact, virtualisation actually inserts a new layer in the IT infrastructure which itself needs to be managed and secured. Some case studies and articles mentioned difficulties faced with regard to clarity on licensing, vendor SLAs and also in deploying management and monitoring tools that would work across both virtual and physical landscapes. Another concern mentioned in the same respect was that the performance of applications in a virtualised environment was not certain.

These views about concerns did get to us at times and led us into thinking about what is the right path to take, but we also knew that any new technology has to go through cycles of proving the benefits vis-à-vis the effort and investment. Therefore when it came to deciding whether or not we should take the leap of faith, we went for it as we felt geared up to face the challenge. We were also conscious of the fact that the virtualisation project would need huge focus and efforts to leverage the true benefits of Virtualisation both from our side as well as from the vendor’s side. The primary and the most important stage was when we had to decide our technology and implementation partner as it is one of the most critical factors contributing to the success of a project.

Before implementation a major task that was to be done was to study our own environment and way forward keeping in mind complexities involved in a virtualised environment and segregate the applications based on whether they would continue in a standalone environment or whether they would move to a virtualised environment. I must confess here that the implementation was not a cake walk as was shown during some power point based presentations; rather it had its own set of challenges where there were occasions when we found ourselves grappling over seemingly minor issues on which teams would end up spending nights trying to resolve.  Also, we went through extensive testing of various applications before moving to the virtual platforms in order to ensure that the transition did not affect the day to the day use of these applications by our end users.

The project aimed at achieving both tangible benefits in terms of a promising ROI and a significant step in our Green IT efforts, which is core to our management’s environment conservation philosophy. Despite the concerns and challenges that the project faced upon its commencement, all the efforts paid off well and the project team comprising the internal IT team, our partners and also our key business users ensured that project was delivered successfully. The role of the business users was very significant as they were the ones who were responsible for the testing of the applications in virtualised environment which was required to ensure that applications behaved as expected with proper business logic.

Like any other new initiative, Virtualisation also had to face internal apprehensions regarding whether the benefits it promised would be achieved but today after being on the platform for ample time, we can surely say that we were able to achieve the expected returns on our investment in less than six months and there have been recurring savings on expenses including energy, space and management overheads. Virtualisation has IT to contribute our bit towards Hero Honda's green efforts.

Another important benefit derived out of this initiative is that now the expected downtime of applications in case of any failure has come down drastically and this is an important step in our business continuity endeavours. Finally, we were able to continue within the same data centre till date.

The success of the project built our confidence in virtualisation and this is the reason why we consider virtualisation to be an option whenever we plan for any new application or system.

By Vijay Sethi