New Schneider Electric White Paper Analyses Data Centre Architectures supporting the Open Compute Project
Wed 8 Jun 2016 | Kevin Brown
A new White Paper from Schneider Electric considers the capital costs of OCP-compliant data centres, including the implications for redundancy, availability and flexibility of upstream power infrastructure.
London, United Kingdom – June 8th, 2016 – Schneider Electric, a global specialist in energy management and automation, has announced a new White Paper #228: “Analysis of Data Center Architectures Supporting Open Compute Project.”
Kevin Brown, Senior Vice President of Data Center Solutions at Schneider Electric, said: “The Open Compute Project (OCP) has had a significant impact on data centre design, with the focus to date being primarily at the rack level. However, there are important implications for the provision of upstream power that need to be taken into account to ensure that expected levels of redundancy and maintainability can be delivered as well as the flexibility to support both traditional and OCP equipment.”
The new White Paper addresses three critical questions about electrical architecture to aid data centre managers in the decision making process when considering an OCP adoption:
- How do OCP rack/server designs affect the upstream electrical architecture?
- Can I still achieve 2N redundancy (sometimes referred to as tier 3)?
- How do I support traditional and OCP loads in the same data centre?
White Paper 228 describes cost comparisons between a number of scenarios including a traditional 2N data-centre architecture, an OCP-specific 1N architecture, an OCP-specific 2N architecture and a simplified 2N architecture with the flexibility to support mixed (i.e. traditional and OCP-specific) loads.
In broad terms, compared to a traditional 2N data centre, an OCP-specific 1N data centre offered capital-expenditure savings of 45%, when including MV switchgear down to and including rack and server power supplies. The White Paper shows that the reduction in redundancy and server power-supply differences represents the majority of the savings.
An OCP-specific 2N data centre offered savings of 25%. Of that savings, 14% is attributed to architecture simplification and thus is independent of load type.
A simplified 2N architecture that is flexible to support mixed loads has a small premium of 3% over an OCP-specific architecture. Mixed load environments are most likely to be adopted in the coming years as data centers migrate their end-of-life traditional equipment. The white paper provides a sensitivity analysis to show how the server power supply cost and the load mix (%) impacts the overall architecture costs.