The Agile Approach to
Data Centre ManagementThe Agile Approach to
Data Centre ManagementStarHub Business Site
StarHub Business Site
The Agile Approach to
Data Centre Management
There is no disputing that the nature of computing has evolved vastly over the last decade. Today, x86 virtualization is part and parcel of the computing paradigm, while cloud computing is easing its way into the mainstream, with adoption of the latter pervading into both small and mid-sized businesses and the enterprise alike.
Yet the infrastructure that undergirds both virtualization and the cloud has not kept up with the winds of change. Today, many data centres continue to rely on similar operating and management methodologies as a decade ago. While it makes no sense to change for the sake of change itself, advances in technologies and new software-centric approaches in managing the data centre does offer compelling advantages in both efficiencies and cost.
Before we can gain a more accurate appreciation of the advantages that the next-gen data centre – or Data Centre 2.0 can offer – we need to take a deeper look to better understand about the original purpose and limitations of a traditional data centre.
Understanding data centres
The simplest way to understand the data centre is to break it down into its basic building blocks. This leaves us with thousands of networked servers that requires power and cooling, and which are protected from physical tampering by multiple layers of security controls and human guards.
Though various improvements and breakthroughs have taken place over the years, they mainly pertain to greater efficiencies in power distribution or cooling, or in terms of more robust or sophisticated barriers against physical entry. In a nutshell, data centre operators have benefited being able to reduce their costs with regards to manpower and power consumption, while offering more robust and reliable data centres than in the past.
When all is said and done, the average customers are still being sold empty server racks that are wired with redundant power supplies and cooled, but left with the ultimate responsibility about how to manage or monitor them. This gives them limited visibility into their infrastructure by default, forcing administrators and technicians to make on-site visits for even the most trivial of maintenance tasks.
An alternative solution is to invest in pricier servers with advanced remote management capabilities, or to establish their own set of control and monitoring systems for their IT equipment. Of course, the latter tend to be of limited utility as they are operated separately from the systems managed by the data centre operator.
Taking the software-defined approach
This is where the software-defined data centre comes into the picture. Rather than treating data centres as building structures that simply deliver space for servers and the basic building blocks outlined above, the next-gen approach gives organizations real-time visibility into all critical systems that is in some ways superior to being physically present.
All systems are presented on a single pane of glass, and offers very fine granularity such as being able to monitor the temperature reading at the rack level, viewing of event and activity logs to pin down intermittent and hard-to-diagnose problems, or receiving instant alerts and notifications on the most minute change in the status of servers.
Moreover, storage, computing, network and security are contained in a single stack and managed dynamically. This makes it easy for businesses to scale up or down, allowing anomalous events or technical hiccups to be resolved quickly.
All these are offered by default, this allows administrators to manage their data centres from a mobile device, or from halfway across the world. Interested to find out more? You can learn more about the software-defined data centre here.