In this week’s Voices of the Industry, Jeff Klaus, GM of Data Center Software Solutions at Intel DCM, explores where to look for automation opportunities in the data center.
The promise of automation looms large over many industries in 2017. For many data center managers and IT professionals, automation has already been a major asset to their day-to-day lives. For others, automation remains an untapped resource for increasing IT efficiency and infrastructure capabilities. Regardless of where your data center team may be along the spectrum, below are some best practices for implementing or continuing your automation strategy within the data center.
Jeff Klaus, GM of Data Center Software Solutions at Intel DCM
Start Small and Build
Making the switch from manual to automated tasks doesn’t have to happen overnight, and it’s unlikely that 100% of your processes within the data center will all be automated. Start with concentrated segments of your infrastructure that are simple to establish and likely to yield the largest time or cost-savings. There’s always room to expand, but if done incorrectly at the start, tasks that were supposed to be automated now turn into massive manual time sinks that put the entire IT team on their back foot.
For teams that have already begun implementing automated processes, do you have an accurate, holistic picture of how it’s going? Press pause to evaluate where early investments have seen a positive return and think strategically about how to build out this offering to other segments of your infrastructure that make the most sense.
Speaking of Building…
As data centers continue to sprawl across the world, IT teams bear the brunt of these expansion efforts. Automation has one of its best use cases in this setting – improving times for new servers to be deployed, configuring that hardware with the existing network, expediting software installation, and more.
[clickToTweet tweet=”Jeff Klaus – Making the switch from manual to automated tasks doesn’t have to happen overnight. #datacenters” quote=”Jeff Klaus – Making the switch from manual to automated tasks doesn’t have to happen overnight. #datacenters”]
Automation can also help data center managers maintain control over infrastructure they can’t support around the clock in-person. As the sever sprawl continues to increase and data centers sprout up to meet the demand for more storage, computing power and higher traffic capacities, the ability for a data center manager to manage various data center’s remotely is becoming increasingly important.
When Done Right, Automation Can Mitigate Human Error
That wonderful IT team we keep bragging about performance? In reality, human error is inevitable – especially when it comes to rudimentary or repetitive tasks that require lower attention to detail and leave the door open for oversights. Automation can help mitigate these mistakes, reduce large-scale outages, and free technicians to address higher priority items on the docket.
Listen To Your Team
[clickToTweet tweet=”Automation can help data center managers maintain control over infrastructure they can’t support 24/7 in-person. ” quote=”Automation can help data center managers maintain control over infrastructure they can’t support 24/7 in-person. “]
Whether you’re just starting or well into your automation transition, take the time to engage with your teams and solicit feedback about these changes. Understand where inefficient processes and time-intensive tasks are watering down manpower that could be better spent elsewhere. Determine a realistic starting point or additions to current automation capabilities with input from your team in the trenches. The best implementations of any new program those where buy in happens across the team – ensuring not only utilization but also consistency across the organization. After all, the IT and data center team overseeing the data center is the group with the best understanding of how these systems operate on a day-to-day basis.
Creating an ‘A’ Grade Automation Strategy
Automation is poised to explode across a variety of industries and data center infrastructure management is certainly one of those areas. Enterprises across the globe have already begin to master basic automation projects, look for more complex, robust systems and tasks with the ultimate goal of passing the torch from the hands of IT professionals and data center managers over to automation. The human touch and intervention is still a vital component of successful data center management, and will always be a necessary component to maintaining any data center environment – but automation opportunities in the data center offer clear benefits that will continue to be wielded by the frontrunners of DCIM.
How is your team implementing automation within your data center environment?
Jeff Klaus is the GM of Data Center Software Solutions at Intel DCM.
Data Center Frontier is edited by Rich Miller, the data center industry’s most experienced journalist. For more than 15 years, Rich has profiled the key role played by data centers in the Internet revolution.
Does DCIM reach component failure in critical component of the electrical architecture.? Example in our 3 Tier system key component is ATS.
Incorrect monthly testing MOP did not make discovery early to head off the problem. Now plagued by parts we are in undesired state for last 2 weeks. Since we are the most valuable utility client, we only have planned outages. Ofcourse we introduce faults once in a while when old equipment is not replaced in time.
But back to ATS,. First the onsite tech diagnosis was motherboard failure (matched on site maintenace contractor diagnosis).. ATS was correctly tested with power loss simulation engaging the dedicated generator. During further examination of 2 ATS in the architecture, linear motor that operates the switch was notice to have suspicious viusal disfiguration. The decision was made to replace it. But has hit the parts availabilty snag. GE ZENITH consortium has apparently taken over the ONAN product line above 1200 AMP. Here we have classic business interferance with DC operations. So DCIM should go beyond traditional role but must use AI components for proreaction to business changes too. In matter of inventory of critical parts.