The five nine's of uptime
Hosting agreements have started featuring a demand for 100% of uptime. The reality of this is something debated by many data centre owners and clients. How is uptime measured and more importantly, how can maximum uptime be achieved?
Uptime is a measure of reliability and is used by those in the industry to describe how long the infrastructure has been online and available for use. Furthermore, it is measured as a percentage against timespan which means 100% annual uptime would infer that a business’s infrastructure was working continuously with no outages for one year, something that not all businesses accomplish. The term ‘five nines’ originated in telecommunications and has spread across a plethora of power-hungry industries including data centres.
Outside of data centres and facilities management, most businesses find it challenging to achieve a 99.9% uptime throughout the year. In 2017 one of the social media giants reported almost seven days of downtime which translates as only 98% availability. With this in mind, asking IT managers to achieve a goal of 99.99% uptime or more may initially seem daunting but there are ways of making it achievable.
It is important to remember that ‘the five nines of uptime’ is an industry buzz word by which data centre availability is measured. It refers to 99.999% system availability over a one year period. To put this into perspective it means just over five minutes of downtime across the year.
Rob Mather, Director at Power Control commented, “Although originally seen as a figure that the server industry aspires to, it is not necessarily an appropriate measure of UPS efficiency. The effectiveness of a UPS should not be measured by outage duration, it should instead be measured by the number of outages. An outage no matter how short is still a costly event for any business and owners should be advised to install the highest level of resilience possible to achieve consistent uptime.”
This level of system uptime can be achieved by adding UPS redundancy. Placing a single UPS frame into a system plus system (2N) or similar redundancy configuration creates two separate power paths and ensures no single points of failure. In this configuration, an engineer is able to carry out maintenance without a need to power down the load or bypass the UPS. Both modular UPS systems such as the Huawei UPS5000 and standalone UPS systems such as the Borri Ingenio range can be configured in a 2N redundancy.
Depending on the application, striving to achieve five nines may not be necessary. The difference between the uptime of a four nines system compared to five is approximately 45 minutes. Therefore, a business case to improve availability from four to five nines must exceed the cost of 45 minutes’ worth of downtime. Unless the UPS is supporting an application requiring 24/7 uptime, a lower, cheaper level of redundancy may suffice.
It stands to reason that the more nines added, the more availability and uptime a business will receive. The amount goes from being measured in hours to minutes, a difference that could be crucial to certain industries. Power Control has over 25 years’ experience of delivering UPS solutions in varying levels of redundancy to the client’s specification. Power Control is trusted by global UPS manufacturers to install, commission and maintain their UPS systems and provides solutions that not only meet the business needs of not but in the future too.
For more information please visit www.powercontrol.co.uk, email info@powercontrol.co.uk or call the office on 01246 431431
Alternatively please visit https://powercontrol.co.uk/product-category/ups-systems/ for specific product information or email Power Control’s solutions director direct rmather@powercontrol.co.uk