Story image

Data center cooling maintenance crucial to avoiding a meltdown

30 Jul 18

It’s a scorcher out there. On some days, the sun beams down with such ferocity that it feels like you’re on the verge of a meltdown – and now more than ever this can occur far outside the summer months.

Unfortunately, the hottest days of the year and the meltdowns that go with them can have catastrophic effects for those managing data centers in every part of the world.

At the heart of any functional data center is the requirement to keep hardware cool enough that it won’t overheat, malfunction, or just shut down altogether. This prevents downtime, service throughput, and inefficiencies.

Cooling also requires the most maintenance of any data center system such as power distribution, fire protection, and generators.

Data center infrastructure and service provider Vertiv knows the effects of heat on hardware all too well – their experts are often called out to help customers who have ageing cooling systems that are about to fail, or have failed already.

The company says that the ones on the verge of failure are generally systems that are poorly maintained, or they’re not maintained at all.

While saving money is always on a company’s agenda, it will most likely pay a steep price later on if data center maintenance cuts corners. It’s a case of ‘you get what you pay for’ – data center systems are only as good as their maintenance.

According to Keystone, well-maintained systems perform more efficiently and reliably, reducing the overall total cost of ownership. Ideally, you want to prevent cooling problems before they happen. One of the best options is to create a preventative maintenance schedule.

So let’s take data center cooling maintenance back to basics. Here’s what you can do to avoid emergency callouts and high repair costs.

Know your systems - understand what you’re working with and how it should be maintained. What’s the maximum cooling capacity of that data center and is it sufficient for the location?

Don’t always assume a general facilities business has the specialist skills - regularly check critical systems even if they are being maintained. These include computer room air conditioning units (CRACs), room and rack temperatures, cooling loops and chillers, and how the floor vents are arranged.

Take the time to look at your systems to ensure they are being maintained to the standard you expect. Preventive maintenance can be available as a full service contract, or an inspection-only contract. Take the time to review the terms of these contracts.

Engage with the equipment manufacturers or specialist data center services providers such as Vertiv. They can provide the people, knowledge, and expertise to keep your equipment at optimal performance and efficiency.

Learn more about Vertiv’s data center services now.

Chayora announces a strategic partnership with Sinnet Technology
Chayora, a Hong Kong-based data center infrastructure company, announced that it has entered into a strategic partnership with Beijing Sinnet Technology.
Commvault fully integrates backup with Cisco Hyperflex
Its IntelliSnap technology has been validated to work with Cisco HyperFlex hyper-converged systems without the need for third-party tools.
Huawei continues 5G trials despite ongoing concern
Huawei completed the 5G NR test at 2.6GHz spectrum in the 5G trial organised by the IMT-2020 (5G) Promotion Group. 
Experts comment on record 772mil-user data breach
Dubbed “Collection #1”, the data set contains emails and passwords with over a billion unique combinations of email addresses and passwords.
Top risk facing organisations? Why, it’s an IT talent famine
For some time there has been talk about how the IT industry is crying out for new talent and skills, which a lot of people have glossed over. But now Gartner says it is a harsh reality.
HPE invests in services with new A/NZ execs 
With IT services spend growing in Australia and New Zealand, HPE is appointing execs for software and technology services in the South Pacific.
Inspur’s server delivery to Baidu claims new record
After an urgent request, Inspur delivered a shipment of rack scale servers of more than 10,000 nodes to a Baidu data centre - equating to one server delivered every 2.88 seconds.
LISA Double Access fibre management system to launch at Cisco Live
“In a data centre, the protection of the fibre is key, which is exactly what the LISA Double Access offers customers.”