Effectively monitoring data centres – their systems, applications and servers – can not only help streamline process but also enable quicker responses to failures.
However, many in the industry still have a ‘build and leave it’ type approach, Peter Stevens, ITO delivery director - South Pacific at HP, told Computerworld Australia.
“...many clients are not monitoring enough in the data centre,” Stevens said. “Most organisations really are very reliant on IT systems that are either not set up in a way to be redundant, as in if one system fails another one takes over, or they’re not monitored at the right levels to enable them to be able to respond quickly or to fix up things when they occur.”
This is due to several reasons, according to Stevens, such as the broad range of terms monitoring covers. For example, it can mean having a security guard on-site but also include automation tools.
“There’s a whole load of things in between, so I think that there’s not a depth of understanding in the ability to monitor right down to the lowest levels and many organisations probably think they are monitoring quite well in their data centres,” Stevens said.
“But then of course, an application, for example, can fall over but the server continues running. At the end of the day, if you’re monitoring the server, you’ll never know the application went over until your users start having a problem or your customers start having a problem. It’s at those different levels that many companies are not monitoring enough.”
Stevens said monitoring requires expertise to establish, develop and customise automation systems to respond to events. However, the benefits of monitoring can be vast. Not only can it prevent outages or respond to outages quickly, it can have a direct impact on the bottom line – not responding efficiently to failures can result in profit losses.
Stevens said if IT systems experience unmanaged outages, data can be lost, resulting in an avalanche of further problems down the track.
“You could lose your whole balance of accounts, for example, or lose a whole lot of data quite quickly if a system goes down in an unmanaged way, particularly critical systems,” he said.
Monitoring can also streamline efficiencies, particularly when automation systems are set up to immediately respond to events. This frees up time – staff aren’t required to wait and see an event and take action – and allows staff to be deployed to more value added functions, Stevens said.
Despite the benefits, Stevens said clients are still dismissing the importance of effective monitoring. He recommends clients develop a monitoring plan which encompasses the entire scope of their IT infrastructure – there is no point in just monitoring one aspect, such as the network, and disregarding servers, he said.
“You can get so much from monitoring at any of the individual levels or from the robustness at any of the individual levels, be it the application, at the server, at the network or at the data centres, but it’s really a combination of those things,” Stevens said.
“…ultimately, monitoring, at the end of the day, is about making sure that availability is high.”