Updated on 23 September 2010
Rapid growth and expansion have resulted in organisations having multiple data centres across various locations. These data centres themselves have been expanded or enhanced at various points in time, and operations are a mix of infrastructure, technology, cooling and power solutions. A few data centres are loaded to capacity, while others are underutilised.
Amit Sinha, Senior IT Manager of a large manufacturing company, is facing a similar situation. Being a large manufacturing set up, the management has projected an ambitious growth plan. With this, the focus on IT will be enormous as the management expects the IT teams to perform miracles in terms of supporting the company’s growth initiatives.
Sinha is not just challenged by one aspect, but is swamped by the varied questions and assumptions that he has to take into account before finding the right solution. His main concern is whether he has to think of a big data centre transformation project, which would be a Herculean task at this juncture. If so, another question that arises is whether he has to think of implementing a big project and re-design the data centre architecture that can address the challenges of data centre optimisation.
Another component that needs to be factored in is what sort of innovations does he need to plan for while overhauling the data centre framework? His key cause of concern revolves around addressing challenges related to storage and data which are growing with geometrical progression. Streamlining the organisational data flow, along with operational efficiency with robust IT frameworks is also a big worry. Sinha attempted to get feedback from his peers in the industry to gain an insight into the trends and understand the immediate steps he needs to take to address all of this. He realised that it is purely a strategic initiative.
Amidst the chaos, he wants these two questions answered from experts, which he feels, will lead him in the right direction.
THE BIG QUESTIONS
1) Does Amit Sinha need to re-architect the data centre framework and if so, what solutions should he apply and what should his primary steps be to increase data centre utilisation?
2) What sort of business benefits and assumptions should he have in mind before going through the data centre transformation process?
Here are the answers...
DAYA PRAKASH, CIO, LG Electronics India
Sinha must re-architect the data centre framework to get rid of multiple data centres at various locations. Multiple data centres are also running a mix of infrastructure, technology, cooling and power solutions. A simplified and consolidated data centre would ensure high availability and business continuity. This would also become a building block for supporting the company’s growth. Some broad areas that he must keep in mind while designing the overall architecture are:
Data Centre Consolidation: Zero in on only two data centres after consolidation, i.e., primary and secondary. Keep the primary data centre where the main business centre manufacturing unit is located and the secondary as a disaster recovery site at the remote site, which can help optimise the infrastructure resources, while ensuring high availability in terms of servers and network infrastructure.
Virtualised Infrastructure: Virtualisation of servers and storage could help Sinha better utilise resources with reduced space.
Data Warehousing: To manage structured and unstructured data and evaluation of trends using BI tools helps in long-term archiving or data retention, which is one way to enhance efficiency.
Intelligent and Predictable Systems: Using smart or intelligent software which is capable of controlling Power to Rack, PDU using SNMP alerts and proactive monitoring and alerts on any change to the infrastructure could be considered.
Scalable: The data centre design should have a scalability option and the right-sized UPS/PDU delivering highest efficiency.
Cooling and Energy Efficiency: Structured cabling, fully managed close-coupled cooling that neutralises hot-spots and delivers the highest availability and efficiency.
Sinha can look at business benefits and expect a better RoI using various technologies and tools as part of the data centre transformation. The re-design would definitely bring in energy efficiency and keep the cost of power and cooling low, while reducing the carbon footprint. The automation of operations, along with virtualisation and modernisation of IT infrastructure will drive higher resource utilisation at lower operating costs.
Compliance risk and governance is a key factor where virtualisation comes to the rescue, as it helps reduce the law-compliance risk, while improving agility and governance. Another advantage is operational excellence and easy manageability issues that are addressed with data centre transformation initiatives. With consolidation, managing multiple data centres becomes easy, particularly the virtualised infrastructure, because most solutions come with good manageability tools.
SANJEEV KUMAR, Group CIO & Group President - Business Excellence, Adhunik Group of Industries
As a primary step, Sinha should take stock of the existing data centres, various applications running, various users accessing these applications and available business processes.
Before re-architecting the data centre, he needs to segregate his data centre components, including the data and applications and other infrastructure into three categories: vital, essential and desirable. The most important one should be within his control which will be accessed in-house.
The essential data or application layers could be virtualised. While the desirable ones could be outsourced to a third party. This process could help him increase the data centre efficiency and utilise the resources optimally. For direction, he could experiment with the Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) model for some application such as collaboration services or infrastructure. I also recommend that he outsource to big technology vendors who have the capability to do a detailed study, and re-design as per the need of the company and provide the aforesaid services.
Most often the benefits that the company gains from any technology initiative or overhauling process are relative to the growth path it has set. Sinha’s company is undoubtedly on a growth path and has set ambitious growth targets, amidst tough competition.
Since the growth plan is aggressive, re-architecting the data centre is necessary. He should lay out steps for the growth phases and decide how he should revamp the IT infrastructure to suit each phase of growth. He should remember that the life of a data centre can be stretched to a maximum of 5 to 7 years with frequent refreshes. But beyond this, a complete overhaul or transformation is a necessity, if a suitable RoI has to be generated.
With the buzz around the cloud, which IaaS and PaaS are also parts of, a good cloud strategy for the desirable and some essential components would be a good move. However, Sinha should consider the scalability factor when re-designing the data centre, besides addressing the compliance needs.
RAHUL MAHAJAN, AVP - IT, K Raheja Corp
Sinha’s company has aggressive growth plans. Whether it is organic or inorganic growth, it is mandatory for any IT manager to review data centre strategy periodically. I strongly feel that he has to opt for re-architecting his data centre.
The first step towards its transformation would be carrying out the virtualisation of physical servers, if he has still not done so. Implementing some effective virtualisation tools would help him consolidate his servers; he should consider using the public cloud infrastructure for non-critical applications.
Some of the critical applications could be placed on virtual servers. Besides, the creation of regional hubs of data centres would help him host various applications and enable him to build redundancy, which will also enhance network bandwidth optimisation.
Sinha should make his decisions around the data centre re-architecture based on certain assumptions. If properly initiated, there could be enormous benefits. For instance, assumptions made around RoI and cost saving in terms of floor space, cooling cost, power cost, UPS, etc., could be achieved through a systematic approach.
Entirely overhauling the data centre infrastructure would also help in easy scalability during the business expansion process, as the design is based on the future organisational road map. Another key factor is that redundancy in DCs, DRs and network would ensure optimal response time for users.
FIND US ON TWITTER
FIND US ON FACEBOOK
What do you think is the best career move?