X
Business

How on-demand virtual machines helped to boost university research output

Self-service automation of virtual machines has enabled University of Reading researchers working on subjects ranging from climate change to neurology to analyse data more efficiently.
Written by Danny Palmer, Senior Writer
reading-universityheader.jpg

University of Reading: Whiteknights House and HUMMS (humanities and social sciences) building on the Whiteknights campus.

University of Reading

The very nature of universities means they're often at the cutting edge of research, with academics increasingly reliant on vast amounts of computing power to help analyse data. Researchers often need to process unpredictable workloads, and at short notice.

However, finding a balance between investing in cutting-edge technology and improving efficiency, all while staying within budgets, can be difficult.

At the University of Reading in the UK, computing resources had previously been split across different schools, so the IT department took the decision to centralise it. Five different compute environments and associated storage were merged into one centralised system, with added elements of automation.

"An initiative was put in place to have one compute and one storage solution to solve the complexity of having so many different systems," Ryan Kennedy, Academic Computing Team Manager, University of Reading told ZDNet.

"We spent all our time firefighting rather than providing any support to our users, so we made a decision to implement a system that was easy to use, easy to maintain and put self-service in so our users could do a lot of the heavy lifting rather than us".

The university set out to find an automated data centre solution that enabled agility while remaining simple to use -- all without blowing the budget. After examining a number of vendors, the Academic Computing Team looked at Nutanix Enterprise Cloud and found it to be a good fit.

"Our key objectives were ease of use and simplicity to maintain alongside a licensing cost model. What we found was that Nutanix ticked all those boxes," Kennedy explained.

"All the products we looked at could do it at a technical level; we just needed to ensure it was easy to use for our end users with a nice UI, and easy for our engineers to maintain -- and also not have really high licensing costs and able to save money that way,"

Kennedy described installation of the Nutanix system as "one of the smoothest experiences I've ever had," although getting used to rolling out an automated service did impose a learning curve due to it being so different from the previous operation. Still, it ultimately provided the benefits the university was looking for.

Massive efficiency improvement

"Understanding that way of working was very different to what we've done before: we went from a lot of physical tin and having to put a CD in, to automating -- but we found massive efficiency in doing that," Kennedy said.

The automated system has allowed 5,000 researchers -- examining subjects ranging from climate change to neurology -- to spin-up up their own virtual machines whenever they want, rather than putting in requests to IT to make use of the services -- something that could previously take weeks to set up.

"The researchers have access to compute 24/7 when they want it: they don't have to wait for one of my engineers to provision it for them," said Kennedy.

"It's made it so they can log on, fire up a virtual machine, do what they need to do, get their results and destroy it at the end, rather than waiting for IT when we might have taken 10 to 14 days to spin-up one for them".

The system has been in place for over a year now, and has left the IT department examining how it approaches new technology. For Kennedy, the lesson is clear: to be more adventurous. "We always used to play it safe, got things which had been around for years and were tried and tested. But we didn't try new things," he said.

However, such has been the success of automating data centre services for researchers that the university is now looking to expand into other areas.

"We're now slowly expanding it into our corporate environments. The idea is when there's a demand for our timetabling system, rather than manually building a VM, we've got the automation to spin another web server up," said Kennedy.

"So when in the first week of term we see a 90 percent increase in the use of the timetabling system, when all the students are looking at it, it's using that web-scale architecture -- but having the ability to have that safety net around it in a climate where we have to save money".

The Academic Computing Team was initially wary about shifting the servers to an automated, on-demand service, but now sees it as the future for the university. "It's the way forward," said Kennedy.
READ MORE ON IT AUTOMATION

Enterprise automation starts in IT departments, but gradually pays off elsewhere
A survey of 705 executives released by Capgemini, finds enterprise-class automation is still a rarity, but taking root within IT functions.

Looking to compare all different types of automation? Now you can
Organizations must vet and compare different types of automation based on business needs, but there was no way to do this -- until now. Forrester's Chris Gardner shares how.

What Kubernetes really is, and how orchestration redefines the data center
In a little over four years' time, the project born from Google's internal container management efforts has upended the best-laid plans of VMware, Microsoft, Oracle, and every other would-be king of the data center. So just what is it that changed everything?

True private cloud isn't dead: Here are the companies leading the charge(TechRepublic)
Companies like VMware and Dell EMC are heading up the private cloud market, according to Wikibon research.

Editorial standards