Robots are on the rise. One forecaster, the Freedonia Group, predicts that the use of robots in manufacturing globally will grow nearly 11 percent annually for the next three years. Robots are growing 15 percent in the U.S., says Freedonia, and 17 percent in China.
What do robots have to do with the Data Center? Well, in the same way the manufacturing sector has transitioned from a manual, human-driven industry to one that is automated and controlled by intelligent machines, the Data Center is poised to go through a similar transformation (albeit without the mechanical arms).
The Software-Defined Data Center is a new paradigm in application delivery in which compute, storage and networking components are virtualized and delivered as a service. Software will be able to intelligently combine, customize, and commission resources from the server, storage and networking pools to ensure that applications remain available and responsive. This takes us humans, with our slow, expensive and error-prone ways, out of the equation.
Deploying applications are not easy today. A series of independent provisioning tasks must occur across different silos. Virtual machines, server adapters, storage partitions and network appliances must all be provisioned and then interconnected to build an end-to-end service. This is typically done across different management systems requiring coordinated effort across multiple teams. No wonder it typically takes weeks or even longer – and why – according to a Gartner 2012 whitepaper – 40% of IT operational expenses (OPEX) go towards labor!
Related Resources from B2C
» Free Webcast: Build Better Products by Identifying and Validating Your Riskiest Assumptions
Additionally, components within Data Centers have evolved at different rates. Some are virtualized. Some are not. Also, networks lag behind servers and storage in power and ease of use. Turning on network services still requires painstaking, error-prone provisioning processes.
Fortunately, there are emerging technologies such as OpenStack cloud orchestration software. A project co-founded by NASA and Rackspace, OpenStack enables companies to build cloud-based networks by providing a control layer that sits above the compute, networking and storage devices in the data center. This enables IT managers to easily control those resources as a service through a set of APIs and a common dashboard. Turning on applications becomes simple and automated. This allows vendors to enhance the current OpenStack environment by eliminating the constraints of traditional Ethernet to bring more design flexibility, agility and scale to networks.
Provisioning Virtual LANs on a port is a great example of a task we should leave up to machines! And you can, with new OpenStack-enhanced technologies. Provisioning VLANs through the core of your network is completely eliminated. Services are deployed at the edge only – and you can even automate that provisioning through intelligent systems like OpenStack. You also have much better scale (16 million services as opposed to 4,000) and the freedom to move VMs anywhere that you need them to go whether it’s across the data center or across the country. And even automate the network and storage provisioning to follow the VMs as they migrate across and between Data Centers.
What does automation mean for the network engineers performing these tasks today? Liberation from today’s drudgery and menial tasks! This gives them time to focus on moving the business forward. Rather than trying to troubleshoot an issue across three separate teams, or configuring Spanning Tree Groups device by device across the network, IT can engineer better applications or roll out new ones that enhance productivity or maximize efficiency. Plan for the future. The possibilities are endless.
Granted, we won’t see the likes of WALL-E in our Data Centers anytime soon. However, there will be intelligent software systems – virtual robots – making the job of changing and turning up new services a whole lot easier and a whole lot faster.