Automation won't resolve the data center staffing crisis, at least not yet

These are challenging times for managing data centers that are also opening up opportunities for IT professionals. The lines between development and operations keep blurring.
Written by Joe McKendrick, Contributing Writer

The latest report on data center staffing from Uptime Institute finds skills shortages continue and AI is not expected to reduce requirements in the near future. As demand for digital capabilities grows, a shortage of qualified data center professionals becomes more acute. Plus, today's data center teams increasingly require both operations and development skills. 

Nearly half of owners and operators in the Uptime survey report difficulty finding skilled candidates, up from 38% in 2018. As such, 75% of respondents believe that most data center professionals have long-term job security. Three out of four owners and operators believe artificial intelligence will reduce their data center staffing needs at some point, but half project this shift is more than five years away. 

In the meantime, industry leaders agree that these are challenging times for managing data centers  --  and challenging for IT professionals as well. For starters, the lines between development and operations keep blurring. 

"As data itself moves to the cloud, and a software-driven operating model, basic administrator roles -- platform administrator, database administrator, network administrator --are being transformed into roles with software-development capabilities," says Arthur Hu, senior vice president and chief information officer at Lenovo. "The new jobs require an understanding of how to leverage the software-based technologies behind today's data center automation. Examples of these types of roles include site-reliability engineers, cloud engineers, solution architects, and others that that drive operational efficiency and cost-competitiveness of data centers."

The challenge with evolving data centers is more about job augmentation "than jobs and roles being displaced as data center automation evolves," says Chris Napier, vice president of operations at CyrusOne. "Contractually, large data centers are always going to require a 24/7, 365-day hands-on approach. A data center engineer now not only has to be able to repair equipment or modify the mechanical aspects, but do it with the technical aptitude of someone who is more or less a developer. It's not a displacement, but an evolution."

With the data center is rapidly automating, "a DevOps skill set that blends the capabilities of application development and infrastructure administration is well suited to this role," says Paul Speciale, chief product officer of Scality. "The key value here is the ability to code scripts, and leverage APIs that are exposed by infrastructure solutions such as networking and data storage."

The use of APIs are a central part of data center automation, Speciale adds. "While this won't necessarily fully displace the storage admin role since traditional storage array solutions are certainly still deployed, it will over time create a shift toward more developer oriented skills for new hiring in the data center."

What opportunities are available for managing automated data centers? "Enterprises will continue to move to the cloud," says Hu. "As AIOps becomes more mature, it will play a bigger role in data center operations, leveraging technologies such as big data and intelligent analytics. IT professionals who understand these fields will have a role in shaping and managing this transformation."

Napier reports seeing "a boom of technology companies and technology platforms, as everyone has some version of a data center management information system that's being developed." This "not only requires developers writing the software and salespeople to sell it in but an engineer who understands how to integrate the software into the data center."

A data center engineer "must still handle preventative maintenance on a piece of equipment but have the tech savvy to use mobile apps on a handheld and be able to dig into any type of system and peel the telemetry out of it. 'Look listen, and feel' is now done remotely through some type of software before the engineer even gets into the room. It goes beyond understanding the structure of data centers to how to apply the software."

Increasing automation "will leverage applications and tools that make use of AI and machine Learning, as well as tools that can analyze large amounts of data," says Speciale. "Opportunities therefore exist for people that can work as data analysts and data scientists, with an ability to leverage newer tools to find insights in vast amounts of data.

For what kinds of jobs should current data center professionals be preparing for? What kind of education or skills should they be pursuing? "Data center professionals can manage the shift to new roles in several ways," Hu reports. "First, they should be familiar with the underlying architecture of the various public and private clouds. In addition, they should hone their skills in critical areas such as Infra as Code, DevOps, AI Intelligent Analytics, data operations, the build and operation of container (Kubernetes), service mesh and serverless. With this knowledge, they will be well-prepared to continue managing data centers now and in the future."

In many ways, "we are looking for a unicorn -- that combination of someone who has the diagnostic thought process on a technological platform," says Napier. "I don't think we'll ever replace the data center engineer, but jobs will be in marrying two opposite ends of the spectrum - the integration of the software and the training of your original technicians, who have been to trade school or are factory-certified with specialties in electrical and controls, HVAC/mechanical and telecom/structure cabling.  It's the mechanical mindset with a data overlay."

Speciale reports seeing "a growing need for skills that blend coding and IT skills -- a DevOps skill set." There is also a new wave of "cloud-native applications and infrastructure solutions that are based on Kubernetes that will be highly valued. These will be the next wave after server virtualization and cloud technologies, so people with early experience and education around open source technologies, Kubernetes and it's ecosystem will be highly valued."

There is also a trend toward an "IT generalist" role emerging, Speciale adds. This contrast with "the more specialized siloed skill sets of the Linux admin, storage admin, network admin of the past," he continues. "The need now is for speed and agility, which requires people with skills across the range of infrastructure within the data center. Since automation implies tooling, scripting and the use of APIs, a blend of developer and general IT skills will be highly valued."

Editorial standards