The Challenge
The legacy data centre was a monolithic block of space, power and cooling. The infrastructure was designed to support large mainframe installations and was managed by a traditional facilities team with little or no input from the IT team.
Data centre usage changed as IT systems evolved to racks of servers that spread the heat and power loads over a much larger area. Although this allowed data centres to scale more easily, it created challenges for power, cabling and system management.
With the ‘one box for one workload’ methodology, IT teams would over provision requirements to allow for future growth. This resulted in a proliferation of comatose servers and unused storage. Server sprawl quickly consumed the available resources and IT managers were forced to exceed the design limits of their data centres in order to meet the ever increasing needs of the business.
The Solution
Virtualisation offers a solution by taking traditional IT hardware and turning it into a shared resource where servers and storage come together into a single service.
The virtual data centre has many benefits, including better resource utilisation, greater flexibility and easier management while reducing your costs.
Do more with less
When you have multiple services sharing resources, it can reduce the requirements for your IT services by increasing the utilisation of your hardware. Resources such as disk, memory and processor can be dynamically added or removed as required.
In your virtual data centre, you don’t need to over provision hardware for resilience because its built into the infrastructure. If a physical host needs attention, the hypervisor can move the guests to another host, usually within the latency of the application which means the end customers don’t even notice. The hardware can then be restored to normal operation without impacting live services.
Pay as you grow
A virtual data centre allows you to create a modular design so you can start small and scale up as demand grows. When you need additional capacity or resilience, you can extend your infrastructure to an alternate site, which could be another office in your enterprise or even a service provider in a public cloud.
By spreading your infrastructure across multiple sites, you don't need to invest in a single large data centre which can help reduce your costs as smaller data centres are cheaper to build and to operate.
You also don’t need to provision additional hardware specifically for continuity. Operating multiple data centres reduce your risks because you can use the capacity at another site to provision standby systems or to backup the live systems. In the event of a failure, you can even dynamically move services between your data centres.
Bigger Bang for Your Bucks
The automation available with virtualisation can reduce the time required to provision services which means IT staff spend less time building servers and more time supporting the business. This makes your IT services much more agile and able to respond to changing requirements more quickly.
Virtualisation also takes away the headache of hardware upgrades. With the IT services no longer dependant on the physical hardware, it can easily be upgraded or replaced making it simpler to take advantage of increased processing power and energy efficiency. That means no more outages for maintenance or downtime for hardware failures.
We have come full circle and are now concentrating more computing power into a single rack. However, "virtualisation" has not repealed the laws of physics, as the hardware is very real, and it requires a lot of energy and cooling resources.
In a virtual data centre, you need to have insight into how your IT services are being deployed to support business objectives in a cost effective manner. Your IT staff have to focus less on the physical infrastructure and more on the services it delivers.