Vasyl -

Investments in on-premise enterprise datacentres on the up as demand for multi-cloud rises

IHS Markit's research into the investment priorities of enterprise datacentre operators suggests cloud is fuelling infrastructure upgrade projects

Enterprise datacentre operators are continuing to invest in readying their facilities for multi and hybrid cloud adoption, suggests research from IHS Markit.

The market watcher polled 151 private datacentre operators in North America to gauge how the move to cloud is affecting their IT infrastructure investment strategies.

While many enterprises talk about winding down their datacentre investments by going “all-in” on public cloud, IHS Markit’s research suggests others are ramping up the amount they spend on their server farms in preparation for greater cloud use.

As such, respondents said they anticipate the average number of physical servers within their datacentres will increase by a factor of two by 2019.

In particular, the organisation’s Data Center Strategies and Leadership North American Enterprise Survey said – rather than tearing down their on-premise datacentres – enterprises are investing in technologies to support multi-cloud deployments.

This, in turn, is leading to the creation of increasingly heterogenous datacentre environments, as server diversity increases, said IHS Markit, and organisations invest in network upgrades, software-defined storage and solid state drive (SSD) technologies.

Read more about datacentre investment strategies

Cliff Grossner, senior research director and advisor for cloud and datacentre at IHS Markit, said these investment areas are laying the groundwork for enterprises intent on adopting emerging technologies and agile software development techniques.

“Application architectures are evolving with the increased adoption of software containers and micro-services coupled with a DevOps culture of rapid and frequent software builds,” said Grossner.

“In addition, we see new technologies such as artificial intelligence (AI) and machine learning (ML) incorporated into applications. These applications consume network bandwidth in a very dynamic and unpredictable manner, and make new demands on servers for increased parallel computation.”

To reinforce this point, Grossner pointed to the emergence of multi-tenant servers capable of running both virtual machines and software containers, which – in turn – is prompting enterprises to demand more functionality from their central processing units (CPU).

“We can only expect this trend to continue and new compute architectures emerging in response,” he said.

The survey also asked respondents for details of the applications and workloads they are continuing to run out of their corporate datacentres, with 30% saying they use their server farms to host “general purpose IT applications”.

A further 22% said they use their datacentres to run business productivity tools, such as Microsoft Office, while 18% said it is where they prefer to run their communication and collaboration applications.

Read more on Datacentre backup power and power distribution

Data Center
Data Management