What is true in business management is also true in IT systems: If you can't measure it, you can't improve it. For infrastructure managers, the ability to achieve uptime and performance can be severely diminished without complete visibility into the state of their infrastructure. What's more, poor visibility translates into more site visits to troubleshoot or restore systems, dramatically increasing operating cost.

The rapid growth of edge computing environments will only exacerbate the emphasis for greater visibility. As the number of disparate deployments increases, the need to keep multiple, geographically separate sites operating at peak performance through ample monitoring will only increase. Keeping these remote sites up and running calls for a comprehensive strategy on key power systems that are prone to failure: one of them being the uninterruptible power supply (UPS).

Power visibility in UPS

Thankfully, the technology exists to monitor batteries in many UPS appliances today. Where it took guesswork, a rigid (and costly) replacement cycle, or outright failure in the past to prompt a battery replacement, the situation is no longer the same today.

Data collected from UPS about their batteries can be leveraged by IT solution providers to better manage the batteries. Using the data collected through a data centre infrastructure management (DCIM) or cloud-based management systems, the expected end of life of individual batteries can be predicated with a high level of accuracy.

At Schneider Electric, we recommend a battery replacement when the threshold of 40 per cent wear is reached on our UPS Score. This means that a battery operating at a higher level, say, 70 per cent, is good for another two years at least. This data is useful for other predictive maintenance tasks, too.

For example, persistently high temperature or humidity might indicate inadequate cooling or problems with airflow within the data centre. Moreover, UPS units can be pre-emptively decommissioned when they are used beyond their design lifespan, or when they start failing internal system checks.

A predictive approach

A predictive approach based on monitoring of UPS batteries allows IT service providers to deliver a host of benefits to users. They include:

• Reduce cost: IT solution providers can schedule fewer site visits, consolidating site visits to perform battery replacements with other on-site maintenance tasks.
• Better reliability: With quantifiable data around the reliability of crucial UPS systems, it becomes easier to prevent downtime. This is particularly important with edge computing sites located at remote locations or the myriad of deployments with no IT employees present.
• Increase in engagement: With no perpetual fires to put out, service providers can engage more actively with customers; customers also gain peace of mind.

Our UPS Score assessment is scored based on anonymous benchmarking against comparable devices, looking at factors such as battery temperature, UPS age, battery age and other considerations. You can learn more about how it is calculated and reflected here.

Remote monitoring of UPS can be performed using the Schneider Electric EcoStruxure IT Expert service, which collects a list of parameters including cumulative discharge cycles and the above data points to calculate battery decay.