White Paper examines impact of Cloud Computing on IT infrastructures

Eaton has published a white paper that explores the latest thinking on powering converged infrastructures, including the growing tendency to believe that power protection and power management are optional, and outlines some strategies for the provision of power in today’s virtualised environments.

  • Thursday, 4th July 2013 Posted 11 years ago in by Phil Alsop

The white paper, entitled Powering Converged Infrastructure and written by Tatu Valjakka, Software & Connectivity Product Manager for Eaton’s Power Quality Division in Europe the Middle East and Africa (EMEA), is available to download from www.switch-on-eaton.co.uk/PCIwhitepaper


Today’s IT environments are typically characterised by unpredictable data growth accompanied by equally hard-to-predict performance demands from users. Data centre operators faced with this challenge are increasingly turning to converged infrastructure to provide them with much needed flexibility at a price that allows profitable operation. When combined with virtualisation, converged hardware infrastructure can also provide exceptional levels of flexibility, economy and resilience. However, to achieve its full potential, the approach to powering converged infrastructure must also be taken into consideration, especially since traditional power architectures are unlikely to provide the solution.


The white paper examines converged infrastructure and the opportunities it creates for novel approaches to optimising resilience. It debunks the dangerous notion that power protection is no longer a crucial issue and offers best practices for powering converged systems, looking at Uninterruptible Power Supply (UPS) requirements in particular.


The paper concludes that for the use of converged infrastructure to provide maximum application availability with optimum cost, effective power protection and management provisions are vital – even when resiliency is built into layers above the hardware layer. The power management system must also be tightly integrated with virtualisation management systems, so that administrators are fully aware of the power status at all times and so that power-fail routines – such as the migration of virtual servers to unaffected physical servers – can be automated. Finally, the power systems must be designed from the outset to deal not only with fluctuating demand, but also future changes in demand, and with the clear understanding that failures can never be totally eliminated.