Hindsight is a wonderful thing. This is why it is important to gather as much information as possible before selecting the products that will protect a critical load to avoid implications further down the line.
Data centres have a duty to ensure that their clients are protected by the highest level of resilience and that equipment is future proofed. But it’s not that easy.
Just like the majority of industries, data centres are under pressure to minimise costs. However, making short-term savings on CapEx can mean OpEx ends up being far more expensive. This can also result in purchasing products that do not offer the flexibility required to support the long-term plans of the data centre. It pays to spend more time considering options from the outset for the best outcome and to reduce the overall Total Cost of Ownership (TCO).
Avoid system oversizing
It’s not uncommon for UPS installations to be designed and configured for a much greater load than is actually required. However, a system which is too large wastes energy, is inefficient and costly to run. It may also cost more than necessary to service and maintain due to its size. Therefore, adopting a right-sized approach means that data centres can literally ‘pay-as-they-grow’, which will pay dividends over the long-term.
True modular UPS systems offer some of the highest levels of availability and resilience. The latest generation of true modular technology is >97% efficient, which can dramatically reduce TCO. The main advantage for data centres is the flexibility and adaptability of the product. Installing a true modular UPS with the ability to use a fully rated frame or empty carcass, provides the option to add the required number of modules to suit the actual load, plus more only when needed. This right-sized approach avoids the need for oversizing, reducing the initial investment of CapEx and minimising running costs. It means investment can still be made in a premium solution and costs controlled at the same time.
Any data centre manager exercising a ‘duty of care’ for the future should consider true modular UPS technology due to their advantages over standalone systems.
Plan for the future
I’ve said it before and I will say it again, adopting the approach: ‘let’s do what we’ve always done’ is not the best practice. It may work but often results in oversized and inefficient UPS systems. Instead, work with your UPS’ historical data to review your load profile. Understanding this will help to size your UPS accurately and create the most efficient and cost-effective system for the future. All the information is available.
Once you have established usage there are three areas of focus; adequately sizing the UPS to the load is the most important consideration. Secondly is to ensure that the UPS is flexible enough to adapt and seamlessly grow with any future load increases. Finally, see if it is possible to utilise the existing infrastructure, avoiding costly changes.
For example: the existing infrastructure may be designed for a 500kW UPS system, but the actual load is only 200kW. A good fit would be to install a 500kW modular frame and either 4 x 50kW modules (200kW N) or 5 x 50kW modules (200kW N+1). This ensures headroom for future growth, reutilisation of infrastructure and optimal performance for the load.
Don’t forget about preventative maintenance
When buying a new UPS, the serviceability of any product should also be considered. Can preventative maintenance and the replacement of essential components be completed without putting the load at risk? True modular UPS systems designed with safe-hot-swappable components may offer a solution. By utilising this technology, data centres have the ability to purchase a system that removes the need to switch to external bypass or reduce the resilience during maintenance, protecting the load indefinitely.
As with any purchase of equipment, a UPS comes with a warranty, this guarantees that the data centre is fully covered for 24 months. The warranty covers electrical faults and component failures. However, the specifics of the warranty may vary from manufacturer to manufacturer. To ensure that this warranty remains valid, data centres must adhere to the manufacturer’s maintenance guidelines.
A standard requirement for most warranties is a minimum of two preventative maintenance visits per year conducted by the manufacturer’s authorised service engineers. These approved engineers can guarantee access to the technical support from the people who designed and built the UPS, plus spare parts and firmware updates.
Over the past decade we have seen the use of Lithium-ion (Li-ion) in the UPS industry become more popular. This advancement enables the reduction in footprint, an increase in operating temperatures and longer operating life. However, Li-ion has yet to replace the use of the tried and tested option of VRLA batteries fully.
When looking at battery options, it isn’t just technology that can help data centres reduce the overall total cost of ownership. Evaluating the required autonomy based on individual power protection plans is essential, not only to ensure there is adequate time to enact the plan but to also prevent oversizing of battery systems.
Ask: ‘how long will it take for the generators to be available?’ and in the worst-case scenario: ‘how long will it actually take to perform a graceful shutdown?’ This information can help answer the question: ‘do we really need to purchase this much lead?’ This in turn saves on budget. Less lead is better for the environment too.
There will always be a drive for data centres to try and minimise upfront costs. However, without careful analysis it could cost the data centre down the line. Data centre managers who are committed to minimising TCO and maximising availability will do well to invest the time to understand the facts, figures and data to help them make the most informed decisions from the outset.
Inviting manufacturers into the discussion earlier to pool knowledge resources, ideas and come up with workable options will also contribute to savings over the long term.