How mature are local cloud offerings?

Three enterprise users share their datacentre experiences with Ulrika Hedquist

The increase in the choice of local datacentres should mean more competition and better prices for users, but are local cloud offerings mature enough yet? Three local users share their datacentre experiences., the official online media company of the local real estate industry, is still, for the most part, buying its own hardware.

“We are looking at cloud offerings, but at the moment, most of our hardware is bought by us because I don’t think the pricing in New Zealand is competitive enough to go cloud,” says COO, Phillip Dunn.

The company has its own rack hosted at ICONZ in Auckland city.

While is currently running a “hybrid model”, using some cloud services, Dunn is not anticipating shifting more systems and applications to the cloud. The biggest hurdle is cost, he says.

“In the US, it’s almost a no-brainer to go cloud – the cost is so efficient, whereas in New Zealand, it’s still too limited,” he says. “There may be cloud offerings but there is no real redundancy. It’s still in its infancy here.”

But going to an overseas cloud provider is too risky for the website, he says.

“Most of our business is in New Zealand. We can’t host outside of the country because it’s just too slow.”

The Pacific Fibre cable was “something we were hanging on to here”, he says, “thinking it could bring some competitive pricing and some true resiliency to the New Zealand infrastructure, but that is obviously not going to happen”.

More local datacentres would be good news to Dunn, because that would “hopefully” bring more competitive pricing for the cloud solutions they offer. But for now, will continue with the hybrid model. Dunn is about to embark on another investment round in hardware, which he hopes will be his last.

“In three years’ time, New Zealand [cloud services] should be there.”


Video-conferencing software provider Faceme offers its customers a cloud-based solution, an on-premise option and a call centre model. The New Zealand-owned company has resellers and staff around the world and is now moving into the global market, says Faceme COO, Richard Rogers. However, the majority of users are in New Zealand and Australia.

“We’ve got Faceme infrastructure in New Zealand and Australia, and we are currently setting up presence in Singapore, Canada and the US,” Rogers says.

The company has been using Maxnet’s datacentre for a number of years, but since Australian company Vocus acquired Maxnet earlier this year, Faceme has also been able to leverage the existing relationship for expansion into datacentres and communications services in other parts of the world, says Rogers.

“We are putting infrastructure into Singapore – we’ve got a partnership there with Vocus’s Singapore datacentre,” he says.

He says that Faceme has the confidence in Maxnet to put a business-grade environment in place. Rogers says the question of data sovereignty has not been overlooked – it is “a consideration and something that we are always aware of”.

“It’s been a consideration for a long time. Whenever you offer software that is outside of a premise, it’s a consideration for customers and for the suppliers,” Rogers says.

Faceme offers in-country datacentres for local presence where it is important, he says, while some of its customers opt for an on-premise model.

“Whenever you put a presence in a country, and you are standing up infrastructure, there are cost considerations to that,” he says. “We are making assessments on that for every country we go into.”

Communication is very important for making the relationship with your datacentre provider a successful one, says Rogers.

“The more visibility and planning you provide, the far less likelihood of any surprises and issues. We have managed to keep our communication strong on both the executive and operational level,” he says.

Ryman Healthcare

The February 2011 Canterbury earthquake destroyed Ryman Healthcare’s office and computer room, giving the IT department a startling wake-up call. After considering its options, the NZX-listed aged care provider decided to move to an infrastructure-as-a-service model in Revera’s Christchurch datacentre, located in the Addington Business Park.

The move was completed in July this year. Thankfully, off site backup tapes on Auckland-based Plan-b’s server platform meant that Ryman Healthcare wasn’t completely disrupted by the earthquake, says the organisation’s CFO, Gordon MacLeod.

“However, post-earthquake we faced the choice between building our own IT infrastructure, or paying someone to deliver it as a service. We’d been talking to Revera for a number of years and in late 2011 chose them as our IaaS partner.”

After crunching the numbers this option worked out slightly cheaper than the in-house option, he says. But at the end of the day, it comes down to how you want to run your IT, rather than cost, he says. The earthquake was an opportunity to start afresh.

“We either reverted to the way we did things or went for something totally different.”

The transition of Ryman’s IT environment to Revera’s datacentre was completed in a month, he says, and all the while Ryman’s national IT network was up and running – supporting 24 retirement villages and eight construction sites.

In addition to VCare, a specialist retirement village application; Greentree financial software; Time Target staff rostering and PayGlobal for payroll, the organisation also runs design and construction applications, as it designs, develops and constructs its villages using in-house resources, says MacLeod.

Ryman also provides internet access to its residents at cyber cafes in each village, he adds.

“All our data is confidential as we are involved with aged care delivery, and we are NZX listed,” MacLeod says. “It was important to us that Revera is able to host our data in a totally secure fashion.”

Another reason for selecting Revera was its disaster recovery service – Silver-lining DR. This is a hybrid cloud that allows Ryman to activate nominated servers and applications at another Revera datacentre if needed, MacLeod says. The primary production systems at the Christchurch datacentre are replicated at Revera’s Hamilton datacentre. So, if a local disaster disables the Christchurch datacentre, the communications link switches to the Hamilton production node, he says. The solution has a 15-minute recovery point objective, meaning that 15 minutes is the maximum time period in which data might be lost in the event of a disaster.

One of the benefits of the shift to IaaS is that it allows the company to focus more on applications, says MacLeod. Letting a specialist provider manage the infrastructure also takes the pressure off trying to stay up-to-date with the latest developments in hardware and infrastructure, which is hard, he says.

“At short notice we can dial up resources, without having to authorise additional capital expenditure and manage commissioning and integration. It’s a big gain,” he says.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.
Show Comments