It’s an interesting time for the datacenter. With the rise of big data and the growing importance of business analytics, the desire to increase cost efficiency and leverage new technologies like OpenStack, and the need for more compute power, the datacenter is back in the spotlight.
At the same time, we have Gartner talking about bimodal IT, where they suggest that organizations should respond to the proliferating demands on IT by adopting a combination of old-style and modern IT practices; essentially advising an enterprise company to operate two IT organisations in parallel. Gartner’s concept of bimodal IT prompted an article in CIO Magazine entitled “What Gartner’s Bimodal IT Model Means to Enterprise CIOs”, which argued that this isn’t necessarily a sensible solution. And at Bright we agree. There is a way for IT organisations to harness the power of traditional IT and experimental / agile IT, and for it to co-exist and interoperate seamlessly. It can be done; we just haven’t been asking that much of our data centers in recent years. So, let’s make our datacenters more dynamic; more powerful; more useful!
We think the answer lies in the supercomputing industry. There are many lessons that Enterprise IT can learn from supercomputing that can help their datacenters be more agile. Underneath any company’s big data and cloud infrastructure lives a huge compute environment. Emulating the best practices and learning from the challenges that have faced organizations running supercomputers will put the modern datacenter in much better shape to face the onslaught of new technologies.
We’re not alone in this thinking. In fact, the next OpenStack Summit (April 25-19 in Austin, Texas), will include new track dedicated to exploring where the cutting edge of the cloud meets the cutting edge of science. When the OpenStack Foundation recognizes that important lessons can be passed from supercomputing to the enterprise, it’s time for CIOs to sit up and take notice.
I recently hosted a roundtable discussion at an IT Leader’s Event, where we discussed the growing pressure on the datacenter to respond to market demand. It was encouraging to hear that organizations right across Europe in the manufacturing, finance, automotive and other industries, are considering a dynamic datacenter strategy to gain competitive advantage. This is the way that companies should be planning their datacenter’s future. And here at Bright, we believe we can help.
In a dynamic datacenter, compute resources can be created and/or provisioned based on workload demand, and in accordance with configured policies. Compute resources may be physical, on-premise, virtual - hosted in a public or a private cloud, or a combination of both.
In all cases, the resources are dynamically created and/or powered on and provisioned on-the-fly for a specific workload. This results in an agile datacenter that responds quickly and automatically to changes in workload demand, reducing power costs along the way.
This is the way that companies should be planning their datacenter’s future.
To discuss this topic in more detail, please do not hesitate to get in touch.