Riello UPS: Always connected: data centre power supply – Leo Craig




Riello UPS provides uninterruptible power supply (UPS) to data centres: a crucial service for banks in an age of 24/7 connectivity where downtime is simply not an option. Future Banking speaks to general manager Leo Craig about why the company's UPSs are so superior to the competition and the challenges of balancing resilience with sustainability.


Could you give a brief overview of what Riello UPS offers solutions for?

Leo Craig: We manufacture uninterruptible power supplies (UPSs), from small ones suitable for minor home office applications, right up to great big ones that will support larger corporations, banks and other critical applications. We are the third-largest UPS manufacturer in Europe and the fourth-largest globally, and we are the largest UPS manufacturer that purely does UPSs - that's our specialism. Anywhere there's data, you need protection of the data and you're going to need protection of the power. That's what we try to guarantee.

What would you say are the challenges in providing an uninterrupted power supply to big banks?

Banks are bound by various commitments to maintain a service, whether it's the data that's supporting the cash tills, or trading desks, and if they were to lose that power, there would be serious consequences for them. What drives our solution is to make sure it can virtually never go down. We can put the correct power infrastructure in so that your data centre will be up and running continuously. What we do is put parallel redundant UPSs in; we can have separate feeds, so you can have two systems, so you never come off power.

It basically makes the system more complex and more expensive, but you're buying an insurance policy and the more you pay for the insurance the better the policy. And I'll always tend to relate it back to that, because people ask, "Why do I need all that protection?". It's insurance; you're guaranteeing that if something drastic happens, if there's a failure, that everything's fail-safe and you can continue running and maintaining service for the client.

What can you tell us about how you develop equipment solutions to ensure this maximum availability?

That's the key: maximum availability and maximum scalability. And it also comes into efficiency as well; because when you're putting resilience in you're putting elements in that basically eat power. So the trick is, when you're putting the system in, to keep the redundant elements as small as possible but still give you that full resilience.

We've developed a product that is basically a modular UPS, which is scalable in modules of 42kW, right up to over 1MW and it's scalable in small sections, so you can scale up and scale down as the data centre increases or decreases in size.

But a UPS is an electronic device just like any other and requires regular maintenance to ensure it operates continuously at optimum efficiency before any potential faults occur. We plan for things going wrong and we plan the system so that it's maintainable; you can maintain the system while still maintaining service to the client. Actually, you can have a fault, still maintain the system and still maintain service to the client, and that's the key element to it. We keep it at maximum criticality, maximum resilience, but maximum efficiency as well, because obviously electricity is expensive.

Banks have to not only meet all the financial requirements, there's got to be corporate social responsibility as well to meet the environmental requirements and save carbon, reduce power and be efficient in terms of power usage.

How do you help banks that are particularly concerned about that angle?

It's identifying the criticality of the application, identifying what they need to do to make sure they have a resilient and robust solution in place, and then applying the efficiency element that is keeping the redundant aspects to a minimum so that you're not wasting energy.

The problem is always that the more resilience you put in, the less efficient the solution will be. But that being said, with the latest product, now we can put that resilience in and they're more efficient than what older solutions would have been, so it's less efficient than the most efficient solution, but it's more resilient. It's like a juggling act: you increase the resilience and your efficiency goes down, if you make it more efficient you can end up losing some of the resilience. It's about achieving that balancing act with the efficiency, resilience and operating costs.

How is it possible to achieve that balancing act?

You'll never get a perfect system that is very efficient and very resilient without a huge amount of capex, because of your front-end costs. There comes a point where it becomes financially unviable to try and achieve that utopia of ultimate efficiency and ultimate resilience, so there is that settle-in point. But that's what we work with the clients to do, is to say: "Look, that's what you can do, this is what's best practice; if you spend a bit more we can actually increase that resilience but is that giving you the benefit that you actually need?"

What's your approach to customer relations?

Normally, when the banks are getting involved, they've often got their own consultants who will advise them on the overall data centre infrastructure and electrical infrastructure. So not only do we work with the banks, we also try and work with the consultants as well, to say what can be achieved in the marketplace today and the best way to do it.

Because consultants can't keep abreast of everything, because they're looking not only at the UPSs but also at the air conditioning, they're looking at all things within the structure - they can never be on the front foot in every discipline within the data centre. It's up to us to talk to the consultants and tell them what they're trying to achieve and how we can help them achieve that for the bank. This helps them with their own qualifications, so we actually help the consultants as well, directly with the bank or independently, keeping them up to date with what can be achieved with the latest power solutions.

Is Riello UPS 'provider network agnostic' and, importantly, is your equipment agnostic too?

Buying a UPS system is like buying a car; basically, you buy it and you've got to maintain it. It has to be serviced every six months to make sure it is all working tip-top, as a correct resilient solution should be, and checked over to keep it running - the same as a car.

The problem is that a lot of the UPS manufacturers actually keep the keys to the device, so that only they can maintain it. You can't go out to any other competent body to say "please maintain my UPS"; you have to go back to that manufacturer because they've kept the keys. It's an electronic key, which means that if something goes wrong, you have to call them in, and they'll put it right because they've got the electronic key and no-one else can touch it.

All Riello products are open protocol, which means that customers aren't locked out in this way; you can go and get any competent engineer, or company, to maintain the Riello UPS. We've actually trained over 300 engineers, attached to multiple companies, so customers can go 'open tender' to get the best maintenance prices from Riello-trained engineers that are working for different companies.

They can go to a non-Riello-certified engineer as well, if they want to, but if you've got a critical application, you want to make sure that the people you're employing to maintain it actually know and understand the equipment, so you'd like to think they've been trained on the equipment they're going to work on.

We've got a certification process where every engineer has a unique number, and every certified engineer and every certified service partner has a unique number. You can go on to the Riello website and check that number, and that will tell them that the engineer is up to date with training, what they're qualified to do, whether it's commissioning, installation, maintenance or full repairs. So we're actually trying to empower the clients to try and check the people they're using.

Riello UPS general manager Leo Craig.
The best data centres require a non-stop power supply.