Monday, 13 June 2016

Top 3 causes of data centre downtime


Top 3 causes of data centre downtime

By Jack Ward, MD of Powermode


Data centre downtime is costly – and the costs are rising exponentially. Globally, the average total cost of data centre downtime is today pegged at around US$ 9,000 (R140,000) per minute. Unfortunately, avoiding data centre outages is not always straightforward.

(1)  Human error
Human error is responsible for 48% of all data centre outages, according to a recent industry-wide survey. A lack of training, knowledge and experience is blamed for most failures. Miss-matched staffing structures with operational goals and the failure of employees to take ownership of tasks, systems and process are key problem areas.

One of the best solutions for proactively preventing human error-induced downtime in the data centre is automation. Although automated systems are designed and built by humans – and are this prone to errors themselves – routine tasks, including the configuring and management and monitoring of crucial systems, are best achieved through automation.

(2)  Cybercrime
Cybercrime is now the fastest rising cause of data centre outages. This is according to the Ponemon Institute. The research group says cybercrime is responsible for 22% of all data centre outages.

Increasing complexity in the data centre has been a boon for cybercriminals who have taken advantage of this to ramp up their attacks. One of the more obvious counters to this is to beef up monitoring systems. Perhaps surprisingly, many data centre managers rely on employees to monitorvital data centre processes when automated systems are far more advantageous. 

(3)  UPS failures/ power outages

According to reports, nearly 30% of all unplanned data centre outages are due to power outages and the failure of uninterruptible power supply (UPS) systems. Importantly, 55% of a recent US-based survey’s respondents claim that UPS battery failure was at the top of their list of causes for data centre outages. 

As power outages and disruptions are common in South Africa, UPSs have a vital role to play. UPS equipment failures can be dramatically reduced through more intelligent design incorporating standby systems purpose-designed for ‘cycle use’ or lengthy, repeated power outages which characterise the local landscape.

Central to these systems is a purpose-designed long-run battery pack which comprises a number of sealed deep-cycle batteries. Such a battery pack can often be retro-fitted to existing UPS systems. One of the major benefits associated with the Q-on LR (long-run) battery packs is a three-year guarantee - unprecedented in the battery industry.


Integral to this system is a state-of-the-art, computerised battery balancing harness that automatically monitors the state of charge of the batteries and reports to users - via a GSM cellular network - on a wide range of parameters associated with individual batteries in the pack. Potential failures can be identified and addressed immediately, before a catastrophic, costly and damaging impact on the datacentre can occur.

Find us:

  

1 comment:

  1. I am amazed by the way you have explained things in this article. I am really impressed by the way you detailed everything. It’s very informative and you are obviously very knowledgeable in this field. Thanks for sharing such an amazing post. efficient cooling systems.

    ReplyDelete