To optimise the profitability of a data centre, running costs need to be minimised. Beyond selecting the most efficient UPS available, and these days the market leading systems can run at >97% efficiency even at low loads (<10%), what else can be done to reduce overall Total Cost of Ownership (TCO)?
Generally, we find that a vast majority of UPS installations are significantly under-utilised. Most were originally designed and configured for a much greater load than is actually being protected. The problem is that a system which is too large wastes energy, is inefficient and costly to run. It will also cost more than necessary to maintain due to its size. Therefore, there is an argument for constant right-sizing and a pay-as-you-grow approach, to ensure you only pay for what you need. However, the ability to be flexible is also important, as it can help minimise costs and therefore also contribute to reducing TCO.
We’ve all heard of the concept of ‘hot desking’, where company resources are agile and can be moved around the infrastructure. Staff can be deployed where and when needed and occupy existing desk space as required. This takes advantage of infrastructure already in place and ensures resources are physically where they need to be at any given time.
Co location data centres and other multi-site operations also have the opportunity to re-allocate resources, in this case modular UPS equipment can minimise capital expenditure and ongoing running costs. A co location data centre selling compartmentalised space or rows or racks or kW/Sq. foot may shrink with one client and grow with another. That’s the nature of business! When this happens, it is possible to re-allocate UPS modules depending on the load and where the space has been sold. Standard frames which can accept the same physical size and power rating of UPS modules can ensure that if one client decides to reduce their rack space or moves on, the over and above hardware can be re-utilised elsewhere.
In reality, we often find that if a data centre has lost a client or a client reduces its rack space, the UPS equipment originally purchased to support that clients anticipated load, either never reaches full capacity or is switched to eco mode (internal static bypass). Although the energy consumption being used is minimal you still need to consider the costs of running this equipment and the facility still needs environmental control.
On the other hand, there are examples of best practice out there. We recently worked with a data centre that had adopted a completely decentralised strategy, creating a group of small data centres on one site with a standardised infrastructure. This included each data centre having the same type of modular UPS and standardising the rating of UPS frames and UPS modules so that they can be re-prioritised dependant on load demand. This approach allows the modules to be redeployed between the data centres based on priority, to ensure load protection and redundancy is kept at the most critical locations. It also means, if a system needs to be maintained modules can be ‘hot swapped’ avoiding any risk to the load (zero downtime!). This is an example of agile critical power solutions supporting an agile infrastructure environment. Availability has been maximised while at the same time minimising costs.
It’s not just data centres which can benefit from being agile. If a commercial organisation was to close several offices or branches and re-centralise its data storage, being able to re-use assets by relocating modules is a cost-effective option. With edge data centres developing, the ability to move UPS modules to where required has its advantages too.
However, this ‘hot desking’ concept relies on modular UPS systems that are robust and of high enough quality to withstand being shipped and moved around regularly. Some less quality UPS systems are modular but are only designed with modules that are easy to swap and repair, rather than relocate and re-utilise, so keep this in mind.
As well as needing standardised frames and a robust, quality modular system, it is also necessary to manage the whole network closely. A helicopter view of the facility is needed, where the environment can be monitored and manged in detail, to optimise the use of hardware.
This can easily be achieved with a SNMP network management card and some training! Many an organisation will invest a significant sum in a state-of-the-art UPS solution but the £200 SNMP card is probably the best £200 they will ever spend, as long as its plugged in.
The card is an industry standard protocol providing a simple tool to enable real-time management of the system. It identifies any issues and can be used to notify when elements need maintenance, it collects data from the UPS, provides status information and can execute user-defined events such as server shutdowns. In essence, it bridges the gap between the UPS manufacturer and the IT team allowing the client to manage the system to optimise its running efficiency. For clients, this helps them further understand their own equipment, so they can adopt a managed approach to fully optimise their systems.
To make the changes necessary for a data centre to become agile in this way, it will require careful planning of infrastructure. For data centres looking to switch over their existing infrastructure/frames, a TCO calculation can be made to show the costs savings of this move.
At CENTIEL we have a different approach and our team are always focused on how to improve TCO for clients. Implementing infrastructures that offer the ability to pay-as-you-grow and to re-deploy modules according to load requirements will require some intelligent analysis. However, with expert advice from trusted advisers who have decades of experience implementing the most efficient solutions, it is possible to have the highest level of power protection in the most cost effective way. Combine this with SMNP monitoring, developing a flexible system approach and keeping up with maintenance to make better use of the resources available, it possible to reduce costs significantly.
Article featured in Mission Critical Power Magazine March 2020