The Automated Data Center: Two Layers of Technology Innovation
December 17th, 2013 By: Bill Kleyman
This week Data Center Knowledge presents a three-part series on data center automation and the potential role of robotics.
There’s an interesting conversation taking place that revolves around automation, robotics and the future of the data center. We helped jump-start the discussion in May with The Robot-Driven Data Center of Tomorrow, and this week we’re going to look at what’s happening in the world of data center automation and how robotics may make an impact.
Data centers will only become more critical overcoming years. As more users utilize content delivered directly from cloud resources – the data center will need to be able to handle the influx of new demands. This means creating efficiencies at all levels within the data center. And so, we’re seeing automation happen within the modern infrastructure at two layers:Logical and Physical Automation.
- Automation at the logical layer. Virtualization, cloud computing, and the modern data center are intertwined to deliver some pretty amazing workloads. In an ever-connected world, automating workflow at the logical layer is absolutely crucial. Why? This is the only way to dynamically control user influx, new types of cloud content, and a new way that organizations interact with the data center platform. A few examples of automation include technologies like provisioning services. Platforms like Citrix’s Provisioning server or the Unidesk infrastructure are able to connect directly into virtualization brokers to help the delivery and control of both desktops and applications. Other platforms like CloudPlatform, OpenStack and Eucalyptus further help automate and create true cloud orchestration. Organizations are able to granularly control hosts, clusters, various zones, and even core virtual machine resources. Then, we have other technologies which help create even further IT automation and configuration management. Solutions like those from Puppet Labs allow administrators to create a unified approach to automation. Under this type of umbrella, an admin can manage a completely heterogenous infrastructure. This means controlling platforms like VMware, Amazon EC2, Juniper Networks, Google Compute Engine, and even bare metal systems. Furthermore these tools allow for organizations to enforce security and compliance policies by defining the desired state of your system and automatically monitoring all changes against that baseline.
At the logical layer, this level of automation and orchestration will continue to advance. More logical systems are becoming interconnected as the resources they utilize are becoming much more streamlined and efficient. This means that intelligent APIs are removing the amount of hops that applications and data have to take to get to necessary resources for optimal operation.
- Automation at the physical layer. Although entire data center automation technologies aren’t quite here yet, we are seeing more robotics and intelligent hardware solutions appear within the data center environment. Robotic arms already control massive tape libraries for Google and robotics automation is a thoroughly discussed concept among other large data center providers. Furthermore, technologies like those from Cisco and its UCS chassis allow administrators to create powerful “follow-the-sun” data center models where hardware automatically re-provisions itself for the appropriate set of new users. In a recent article, we discussed the concept of a “lights-out” data center. Well, major data center vendors are taking notice. In fact, Panduit is jumping on the automation bandwagon very quickly. Recently, Panduit launched its new Industrial Automation Advisory Services. These new services bridge the gap between IT and Control Engineers for connecting, managing, and automating industrial networks and control systems. The news release goes on to explain that today’s industrial organizations are driven to increase production and reduce costs while maintaining quality and safety. As networks converge, the physical infrastructure becomes even more critical to support the demands of real-time control, data collection, and device configuration.
To paint an even clearer picture, according to Gartner Research, 80 percent of mission-critical outages through 2015 will be caused by people and process issues, and more than 50 percent of those outages will be caused by change/configuration/release integration and handoff issues.