Australian transport supply chain services provider IFC Logistics had long had all its information technology bundled into a self-managed datacentre, but a series of serious outages led it to move to a fully virtualised and automated managed service.
The company’s core business is freight forwarding and third-party logistics, meaning close to 100% uptime is crucial.
Phillip Beal, project manager at IFC Logistics, is responsible for all communications and technology within the business. He recently presided over a transition from on-premise IT to a managed service from Advantage Technology Solutions, a partner of hyper-converged infrastructure specialist Nutanix.
Beal inherited an ageing on-premise datacentre when he took on the role at IFC Logistics, and ran into problems with tech outages that affected the business. “We had a data entry system on-site in Melbourne that we had built ourselves,” says Beal.
But the Melbourne site had become outdated and was proving troublesome. “We had a couple of situations where we had huge problems with the IBM storage area networks,” he says. “First, the disks kept dying. We then kept getting controller failures, which were really frustrating. Because this was an end-of-life product, we had to pay a lot of money to stock parts for it.”
Another problem was the age of the staff providing service for the datacentre. Being young, they had a tendency to tinker with the system. “They liked to play a lot and bought a lot of stuff we didn’t need,” says Beal.
Read more about storage virtualisation
Virtualisation brings benefits to the datacentre, but also big changes to storage provisioning and management. We look at the key challenges and how to deal with them.
Veeam’s Unlimited Scale-out Backup Repository feature in version 9 of its Availability Suite will allow customers to create virtual storage nodes from heterogenous media as targets for backup data.
Server and desktop virtualisation brings many savings, but you have to work smart to ensure storage performs efficiently. We survey the top five pitfalls and how to avoid them.
The problems with the ageing datacentre came to a head when there was a major outage in mid-2015.
“One of the controllers failed and we couldn’t identify the controller as the problem at the time. As soon as you had high disk input and output on it, it would shut down the whole cluster and every virtual machine would go offline,” says Beal.
No place for outages in IT-centric business
Discussions with suppliers proved frustrating, but the controllers were eventually replaced and the system was restored.
But the outages had taken their toll on both the board and Beal. “It was probably the one of the most traumatic experiences of my life,” says Beal.
“We are an IT-centric business, so the business stopped whenever we had a failure,” he says. “There were other problems as well. We had federated services for Microsoft Office 365, for example. Because that went offline, people couldn’t log into Office 365 because they couldn’t be authenticated.”
The outages led to a major rethink of how IFC Logistics would structure its information technology systems in the future. The discussion came down to business fundamentals.
“We asked ourselves what we were in the business of doing,” says Beal.
The answer was that IFC Logistics is in the business of providing transport, freight forwarding and global logistics, rather than running datacentres at a granular level.
“It’s too much of a headache maintaining your own datacentre,” says Beal. “It just doesn’t make sense. You have to treat it as a utility like water or power. You have to look at it as an operational expenditure as opposed to a capital expenditure.”
Moving to a virtualised cloud environment
While IFC Logistics was ready to go for a fully virtualised cloud architecture, it was not ready to move into a public cloud such as Amazon Web Services (AWS) or Microsoft Azure.
“We looked at the feasibility of public cloud, but decided it may not be mature enough yet in Australia,” he says.
The decision was therefore made to move to an Advantage Technology Solutions managed service with Nutanix. This freed up the IT department from having to run a virtualised environment with tricky storage area networks.
“It wasn’t necessary for us to see the underlying technologies,” says Beal. “When you are running an IT department and can forget the daily infrastructure, you can focus more attention on the core operations of the business and dealing with customers.”
The February 2016 move from the on-premise system to Advantage’s managed service took less than a day. IFC Logistics has not experienced any problems since the move.
“When you can forget the daily infrastructure, you can focus more attention on the core operations of the business”
Phillip Beal, IFC Logistics
“Nutanix is one of the few things we haven’t really thought about, but not in a bad way,” he says. “It’s this magical piece in the middle that joins everything together. It’s like if someone asked me to tell them about the kernel in Windows, I’d say it doesn’t really matter.
“I’ve had no problems with storage, and that points to a good system. You don’t have to think about it or put effort into it. It just runs in the background.”
What Beal now sees is several terabytes of virtualised storage, instead of the hardware.
“It doesn’t matter what drive or controller it runs on. I just have storage, and I can provision storage to other virtual machines. Beneath that, it doesn’t matter what hardware is there. It’s better to leave the hardware area and focus on delivering IT to the business,” says Beal.
“The biggest thing for us is that the investment in IT has stayed the same, but it’s going to all the right places now we don’t have to have engineers running the hardware,” he concludes.
