The importance, role and functionality of datacentres will change over the next five years, leaving many organisations unclear on how to plan their future datacentre architectures, reports Gartner.
"Over the next five to 10 years most organisations will need to change their approach to previous datacentre strategies used in the last five to seven years, as most of the world comes out of recession and the nexus of forces (social, mobile, cloud and information) affects technology use, says Rakesh Kumar, research vice president at Gartner.
"Historically, datacentres have been viewed solely as service delivery centres in which cost and risk must be balanced. Agility, a critical third variable, will become increasingly important in future."
Agility is the speed at which the IT organisation responds to business needs such as greater use of mobile and social information to market products, or contactless payment systems. The ability of the IT organisation to respond quickly to these changes will be a reflection of its agility.
Gartner says the following are eight areas to consider when developing a datacentre strategy that balances cost, risk and agility:
Start Deploying Processor, Memory and Power Efficient Technologies
The next few years will bring significant enhancements to process architectures and the economics of processor and memory components will change. In-memory computing, where the primary location for application data is the main memory of the computing environment, will become more widely used, helped by ever-cheaper DRAM and NAND flash memory. At the same time, the use of low-energy processors in servers will increase, with the potential to significantly reduce energy consumption costs.
Move toward a Balanced Architectural Topology and Delivery Model
The architecture of systems deployed in datacentres will change over the next five years and the topology of the datacentre delivery model is also changing. The use of cloud and a range of hosting providers will continue to increase over the next few years as many organisations shift their IT spending from capital expenditure to operating expenditure. The boundaries between the traditional types of infrastructure outsourcing — managed hosting, datacentre outsourcing (DCO), and DCO-related services such as remote infrastructure management — are becoming blurred.
Gartner predicts that these markets will converge over the next decade as providers increasingly deliver their services on cloud-enabled system infrastructure.
Invest in Operational Processes and Improved Tools
Enterprise datacentres are centralised and highly critical IT service delivery hubs relying on well-orchestrated operational processes. This involves understanding, documenting and constantly reviewing end-user service levels and mapping them back into the core IT delivery capabilities. The major areas of concern are around change, problem, configuration and asset management.
Looking ahead, areas such as enterprise security, data management and mapping business processes to core IT processes, will become even more critical — and as agility becomes an increasingly important measure of datacentre value, improvements in operational processes are vital.
Integrate Disaster Recovery and Business Continuity into the Core Strategy
With socioeconomic turbulence in many regions of the world and changes to corporate governance affecting many business areas, strong and well-documented disaster recovery (DR) and business continuity (BC) planning is essential for all large datacentres. However, this must evolve from a separate, specialised project plan into an integral part of the overall datacentre strategy.
The move away from a "just-in-case" strategy to making BC and DR a part of continuous data operations will reduce cost and potentially improve agility.
Manage Capacity Growth through Data Analysis
Over the next five years, most organisations will notice a significant increase in hardware (storage, server and network) capacity, extending to network traffic, datacentre floor space, power and cooling. As organisations start developing the next generation of applications based on the Nexus of Forces, there will be a surge in demand for storage and backup systems and new roles, such as data engineers, will multiply. This increasing but nonlinear capacity is a key a force that must be considered in all datacentre strategies.
Plan for Operating System and Application Changes
During the next five years, changes will be made to the operating system diversity in most large datacentres. A steady migration away from Unix onto the Linux platform will take place, the Windows environment will continue to grow and the IBM Z O/S will see expansion in certain geographies and contraction in others.
These changes will be significant, causing disruption to hardware architectures, application designs and service levels. Gartner recommends UNIX migration plans start by 2014 for all applications, except those with extreme uptime, latency and compliance requirements. Other applications with sensitive client information/access can occur later, around 2017.
Make Consolidation and Rationalisation a Continuous Change Programme
Datacentre consolidation has been a topic of discussion for many years, but many large IT organisations struggle to reap the real benefits because they are complex projects spanning many years and often affect a range of applications. Additionally, in large projects, organisational and political issues often complicate matters.
Organisations should position these activities as a continuous change rather than a one-off project. Continuous optimisation of hardware and physical sites will mean that infrastructure and operations groups are more likely to run at an optimum cost level. They are also able to plan changes better, such as shutting down a physical site.
Modernise Datacentre Facilities
Changes in the hardware infrastructure environment have focused a lot of attention on the capabilities of datacentre facilities. The problem lies in the amount of power and cooling that new high density infrastructures require. Most datacentres are not designed for such power increases and the commensurate rise in costs is typically unsustainable when using older facilities.
Datacentre managers must modernise the capabilities of their facilities to handle both the emerging hardware technology and the escalation in energy consumption as a result of projected growth in server volume. Another important consideration is when to deploy software tools that report, manage and control a datacentre's power and cooling elements. Datacentre infrastructure management tools should be considered as a vital part of datacentre management.
Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.