Coates Hire’s chief information officer, Ben Waterhouse, says the company’s Azure migration has not only cut costs but allowed the company to explore new opportunities involving IoT and machine learning services.
Coates Hire has branches in more than 150 locations around the nation, and has more than 1 million pieces of equipment available to be rented.
The migration to the cloud was sparked by key infrastructure approaching end of life. “The original thinking was very much let’s refresh it, let’s do what we normally do,” Waterhouse says.
“Historically we’d taken this view that to move to the cloud we needed to transform, which we knew was going to be a very lengthy exercise with legacy applications like ERPs and the like,” the CIO says. “But we worked through it a different way,” he adds: “‘What if we just lift and shift everything, and then once we’re there we start the transformation journey?’ And the rest is history.”
“The maths all worked out,” he adds. “So it wasn't actually a, ‘We want to move to the cloud, let’s do this,’” he adds. “It was a, ‘We need to refresh our data centres; wait a minute, we’re taking the wrong approach, let’s look at it differently.’”
Cost wasn’t the only factor in the decision but played an important role, the CIO says.
“By the time you factored in the total cost of running data centres, co-location, licencing, hardware, maintenance, support, et cetera, et cetera, it worked out significantly cheaper to leverage public cloud environments — even just doing a lift and shift,” Waterhouse explains.
Over a period of five years, the migration to IaaS is expected to be about 25 per cent cheaper than the alternative of refreshing Coates Hire’s on-prem infrastructure, the CIO says. However, the savings are expected to be even more attractive as the company works to optimise its use of cloud services following the lift-and-shift.
Beyond the TCO component of the move, it was also a question of public cloud giving Coates a platform to realise its digital transformation ambitions, the CIO says.
“It gave us this launching platform to start immediately thinking about things in a new way, starting immediately to leverage platform services, whether it be for IoT, whether it be for machine learning, whether it’s for security,” he says.
Coates Hire has historically been primarily a Microsoft shop, leaning heavily on SQL Server, .NET and Xamarin, Microsoft's mobile development framework, and that was a factor in the choice of cloud provider.
“AWS was quite compelling from the infrastructure as a service point of view, except we couldn't leverage our existing licences to the same level as we can in Azure,” Waterhouse says.
“When we really dug under the covers and started looking at the platform services, things like API gateway, things like hosting web services, being that we leverage that Microsoft stack we were far more comfortable with what was available from a Microsoft perspective than an Amazon perspective.”
Coates Hire formerly previously had five racks in each of its data centres, with the company shifting some 300 virtual machines, a substantial amount of shared storage, and a number of core applications to Azure, including a “fairly substantial SQL Server workload”.
The migration encompassed the firm’s primary ERP, which is a rental industry specific ERP called Baseplan, its data warehouse, its budgeting and forecasting software (Cognos TM1), and a “myriad” of supporting systems ranging from “best of breed” applications to help manage transport to internally built mobile applications and their accompanying integration backends.
All in all it was around a dozen core applications and a “fairly long tail of other smaller bespoke solutions,” Waterhouse says.
Coates Hire approached the project “like it was going to be a really big deal, and that it was going to be really hard, and it was going to take a long time,” the CIO says. “We put a lot of governance and control around it because we were, to be honest, probably a bit scared. We saw it as a very significant move of our core applications.”
As it transpired, with a “fair chunk of help” from Microsoft, the company was able to move almost all of its on-prem workloads in the space of three days.
For the migration Coates Hire engaged in extensive prep work but “in reality we probably over-engineered that,” the CIO admits.
The process primarily involved migrating physical servers and virtual servers to Azure’s IaaS. However, the CIO says that Coates has been able to leverage Azure’s PaaS offerings, such as App Service, sooner than expected.
In parallel to the migration Coates had a number of other projects running, including developing some of its first public-facing APIs. Azure’s PaaS offerings have made the task easier, the CIO says.
An initial API has involved Coates Hire working with 3D Safety Systems. Coates can make internal equipment data available to the company via an API, and 3D Safety Systems can then present that to customers via iOS and Android mobile devices.
Although the Waterhouse sees increased use of PaaS in the future, he expects to keep a number of legacy systems running on IaaS.
The company hasn’t yet shifted completely into the cloud, but the CIO says he has a “pretty aggressive plan” to whittle away at what’s left. There are some complex, terabyte-sized SQL environments that he expects to move into Azure by the end of the calendar year. He says he is aiming to end up with around half a rack at two different co-lo facilities.
For Waterhouse, digital transformation is “all about continuous innovation; what can we do with technology to help drive a tangible outcome for both our customers and our organisation?”
“For me, it’s not a project that starts and a project that ends,” he says. “It’s about continually adding that value. And I think cloud platforms play a part of that, primarily around agility, speed, and taking away what I call low-value-add stuff, allowing me to invest time on higher value-add items.”
“I can leverage services by turning them on, with a minimal amount of configuration, and have my team focused on building the Coates IP on top of that or the customer-facing part on top of that, rather than investing time, energy, in that low-value-add item,” Waterhouse says.
IoT and machine learning are good examples, the CIO adds.
“We can turn those services on in a cloud platform, use them, and focus on the value addition on top of it rather than maintaining, building, and managing a service,” he says.
It’s early days for Coates Hire’s investigation of machine learning’s potential — “We've got a lot of data, and we're trying to look at how we can use that to the best of our ability, and how we can potentially use that to help our customers as well,” Waterhouse says. However, the company is preparing to roll out an Azure-powered IoT data platform across the business, according to the CIO.