Vision Super needed to replace their ageing data centre environment with a modern approach. While the flexibility of public cloud was attractive, costs were a concern.
Vision Super engaged with Data#3, leveraging their partnership with HPE to review their requirements and identify the most appropriate solution. It was determined that an on-premises offering hosted at separate data centre locations paired with comprehensive Project Management services would meet the high-level performance requirements whilst ensuring redundancy.
The skilled and dedicated project manager from Data#3 made the entire process run smoothly, with no disruption to users and no stress to the IT team.
“The project was entirely driven and justified by risk reduction, and a key success criteria was that nobody noticed, and we met that, which meant we made a safe transition. From an IT operations point of view, the team is able to achieve a lot more by having everything consolidated under a single pane of glass, our ability to monitor and deal with things before they become problems has increased. This means we can deliver a better level of service to our user community.”
James Milne, Head of Technology, Vision Super.
Since 1947, Vision Super has helped Australians to plan for their retirement, as well as offering expert advice on insurance and investment options. Today, the organisation manages around $12 billion in assets for 85,000 members, and in 2020 was recognised with a ‘Best of the Best’ award by Money Magazine for keeping costs low and service levels high.
Vision Super’s data centre was due for an update, and the business recognised that the legacy equipment would increase risk if it were not promptly updated. Rather than simply replace like-for-like, incurring a large capital expense, a pay-as-you-go model was preferred.
Vision Super was dependent on legacy hardware, and although vendor support had been extended, Head of Technology, James Milne, was conscious that holding onto an ageing data centre environment would place limits on the business.
“We had a number of on-premises machines, and these weren’t advancing us in terms of additional capability. There was no opportunity for scaling, and in a like-for-like replacement, we would have to provision for the greatest amount we would expect to use over the life of the equipment. Although we are growing, we would have to provision 100% above what we need now to give us five years, which doesn’t add up. We would also have to manage all that capacity,” explained Milne.
Those management considerations were significant, given the small in-house IT team was responsible for supporting Vision Super’s workforce, as well as undertaking a comprehensive digital transformation.
“Most of our staff were office-based, primarily in our Melbourne head office, and around 15% were mobile. The largest group are office administrators who use our admin systems, take contributions from members, provide customer service, or manage investments, and these people are the heart of our organisation,” said Milne.
Complicating matters, the usually predictable business trajectory was impacted heavily by the COVID-19 restrictions, and the small IT team worked tirelessly to support the business through a sudden change in work practices.
“We knew our legacy systems had limits to growth, because when we hit the challenge of suddenly switching to a work from home model, and we had to provision a lot of machines when people were home-based, we hit our capacity limits fairly firmly. We had to very carefully manage capacity to provide adequate performance for our people,” recounted Milne.
Addressing disaster recovery needs was also a consideration for Vision Super. The business has a strong focus on minimising risk, with Milne committed to ensuring that “no matter what”, the organisation wouldn’t miss a beat as it serves its customers.
Like most organisations, Vision Super was keen to limit less predictable capital expenditure in favour of regular monthly costs. With this in mind, the expectation was that a cloud solution would be the most likely outcome. Milne, though, had done his homework, and was conscious that cloud overspend is rife.
“We were coming from traditional, legacy-based machines that were located on-site, so we knew it had to change. The big discussion for us was about a potential move to cloud, and that was the initial assumption, because it is the modern way. We went through the motions of pricing on-premises options but buying new hardware to replace the old was not particularly attractive,” outlined Milne.
Data#3’s proposal offered a third option. Leveraging HPE’s ‘edge to cloud’ GreenLake platform and SimpliVity storage environment, along with Aruba networking, it delivered a cloud experience, complete with an OPEX cost structure instead of capital expense, but with the advantages of an on-premises solution. With research showing approximately 70% of applications and data still residing on premises, HPE designed the platform to give organisations the flexibility and scalability to stay competitive, while retaining a greater level of control.
The inclusion of project management in Data#3’s proposal was something that had not been included in Milne’s initial wish list, but the more he discussed it, the more he felt this would lift the pressure from his team and ensure minimal disruption to Vision Super’s users. While he said that a complete data centre update is a “massive undertaking”, having a dedicated and highly skilled resource focusing on the project made a tremendous difference.
“The solution was delivered without downtime, without any staff disruption, and without any special training needed for staff, essentially there was never a ‘big bang’ involved. We could provision and start moving workloads and conduct a gradual ramp-up over the course of three weeks. Everything was handled in a steady, controlled way, and we had to announce to the business it was completed, because with no disruption, nobody had noticed,” said Milne.
“For me, as the executive responsible, that couldn’t have been a better result. This is the biggest single project we have executed in the last three years. I didn’t lose a night’s sleep, nobody complained.”
It was important to Vision Super that the busy in-house team would be able to support the solution themselves, so gains in manageability were keenly considered. The solution included ongoing support from HPE, and Milne said that the simpler, ‘single pane of glass’ management reduced the burden on his team.
“Our IT staff now have a lot more time for other priorities – there are only three in the group, so any time we can win some time back, it is a good thing.”
Milne noted that, in contrast to their previous pressures when reaching capacity limits, the business actually had more capacity than was needed. His team was able to effortlessly accommodate the needs of the steadily growing organisation.
“Because of the way the hyper-converged systems handle compression, it was difficult to estimate the resources required, and we found that we had in fact drastically overestimated, so that was fantastic, we had our expectations exceeded,” said Milne.
While there was no immediate impact on most users, there was a more profound difference when it came to tech groups such as those handling business integration.
“We have been able to deliver much greater responsiveness in terms of provisioning new environments. They call us, and by the time the phone call ends, we have done what it used to take three days to do. The new environment has given a lot more flexibility to those tech teams, and we are expecting it to change their way of working as they make use of that efficiency. We can spin up and pull down virtual machines at less cost. In the past, we would rely on them telling us the urgency level of their request and put them in a queue accordingly. Now, it is done in a few clicks,” explained Milne.
Vision Super has earned a reputation for keeping costs low for members, and this is reflected in the IT team’s efforts to ensure the best return on investment. In keeping with this ethos, Milne scrutinised the numbers carefully.
“When the HPE GreenLake solution we were offered ticked a lot of boxes, and we analysed the cost of cloud, it surprised us. So much so that we got the numbers checked. The analysis was right. Cloud costing is also often unreliable, in that most people find it ends up being more expensive than predicted. We are in the position of knowing we can expect steady growth instead of needing to scale suddenly, we are not dealing with huge fluctuation, and there would not be a great deal of benefit with our current architecture to move to a cloud-based system.”
While Vision Super aims to keep costs low, Milne said it is important to look at the overall cost, including the resource time of the IT team. While the Data#3 proposal was not the cheapest, it offered additional value that others did not.
“We had worked with Data#3 before, and although we talked to various vendors, they won our business with a higher quote than some others because they insisted on pricing in the project management. We preferred the responsibility sit with Data#3, which was very successful, and we are very satisfied with our choice. It was the project manager that really delivered a very stress-free project. The alternative would have meant one of our staff members dealing with the project full-time,” outlined Milne.
“Throughout the process, Data#3 and HPE were very collaborative, they were acting as project managers, which is what we wanted from them. The actual design was fantastic, and it was delivered well. They were very tolerant of us when we had to jump through hoops to demonstrate the solution had the best member value, they kept the quotes open, managed shuffling the resources that were lined up when we had delays our side, and understood the constraints we were under. Data#3 just shone as the clear favourite.”
While the usual HPE performance is evident, that was not the driving force behind the project. With Vision Super’s 85,000 members trusting the organisation to secure their financial future, the responsibility is taken very seriously.
“The project was entirely driven and justified by risk reduction, and a key success criteria was that nobody noticed, and we met that, which meant we made a safe transition. From an IT operations point of view, the team is able to achieve a lot more by having everything consolidated under a single pane of glass, our ability to monitor and deal with things before they become problems has increased. This means we can deliver a better level of service to our user community,” explained Milne.
This risk reduction included a disaster recovery solution that eliminated dependence on a single location, should a man-made or natural catastrophe occur.
“Our office effectively became a branch office, there is no equipment we absolutely depend on, which is a major milestone. Because we opted for a hyper-converged solution, the same applies to either data centre, if something disastrous happens, nobody would notice in terms of going offline, beyond maybe two or three seconds of hiccups. We have not just removed reliance on our CBD office, we have removed any single point of failure on the data centre side,” said Milne.
Having worked with Data#3 on previous projects, the Vision Super team trusted that the right outcome would be delivered. Milne said that he was happy his confidence in the relationship continuing beyond initial implementation was warranted.
“Data#3 has also been very good since we went live, they have absolutely delivered. They were there for us when we encountered some minor issues, owned them, and didn’t step away – I am very happy with the relationship. Their greatest strength is their responsibility, and I value that more than anything. I have the complete sense that they take their responsibility seriously.”
Having seen many peers grappling with cost blowouts related to cloud transitions, Milne recommends thorough analysis and comparison of overall expense when considering a data centre refresh and is pleased that this extra scrutiny paid off.
“Our resource usage is significantly below what we projected, which is the opposite of most reports in terms of cloud migrations, where usage tends to blow out and it can be hard to trust calculations. The biggest deliverables have been uptime, and notably lower operational risk. We have just been through a cyber assessment, and we have mitigated a number of risks with this solution,” recounted Milne.
“It has been noted as a success in the business. If only most go-live experiences were as straightforward and uneventful. We are very happy with the outcome.”
1. HPE Greenlake Edge-to-cloud platform (2021) [Online] https://www.hpe.com/us/en/greenlake.html
Tags: Cloud, Cloud-based Disaster Recovery, Consulting, Data Centre Infrastructure, Disaster Recovery, Hewlett Packard Enterprise (HPE), HPE GreenLake, Infrastructure as a Service (IaaS), Modern Workplace, Project Services, Security