Project CoolEmAll needs you

The EC CoolEmAll project is developing new software tools to help with the efficient planning and operation of datacenters. The project, which began in the fourth quarter of 2011 and will run for 30 months until Q2 of 2014, is scheduled to deliver its first prototype tools in Q2 2013. As part of that process, the consortium is looking to increase collaboration with outside experts from the wider data centre industry; help is needed to test the new software but also to shape the future direction of the project.
The need to improve how efficiently data centres are planned and operated is accelerating thanks to continued high demand for new data centre capacity combined with other factors such as the increased competition for energy resources.
The financial crisis may have dampened data centre demand temporarily, but current projections indicate strong growth ahead. By 2020, it is estimated that annual investment in the construction of new data centres will rise to $50bn in the US, and $220bn worldwide.
By approaching the issue of data centre efficiency in a holistic way - that includes not only the IT, facilities (power and cooling) but also the role of workload and application efficiency - project CoolEmAll will be able to solve business problems that are out of scope for most existing tools and suppliers. It is also expected that this kind of holistic approach will also add to the scientific understanding of the myriad complex interactions that define how efficiently a data centre operates.
The seven members of the CoolEmAll consortium are Poznan Supercomputing and Networking Center, The Toulouse IT Research Institute, High Performance Computing Center Stuttgart, the Catalonia Institute for Energy Research, Atos, 451 Research and Christmann informationstechnik. (451 Research’s main role is to provide guidance on the commercial potential of the project and facilitate cooperation with existing suppliers, standards bodies and other industry groups).
The main aim of the project is to increase understanding around the interaction between IT hardware, IT software (applications and workloads) and cooling systems within datacenters. The main outcome of the project will be the development of a Simulation, Visualization and Decision (SVD) support toolkit to allow datacenter planners and operators to model these interactions. The toolkit will essentially help datacenter planners model the energy efficiency implications of physical placement of servers (and other hardware) within the facility, and model different approaches to cooling, as well as the role played by applications and workload optimization. The first prototype of the SVD Toolkit is scheduled for Q2 2012. (The consortium will make elements of this prototype available for review by potential early adopters and partners.)
While the project is funded and conducted in Europe, it is important that it also integrates with international organisations including potential users, standards bodies and regulators. The project has already begun this process of engagement with both European and international stakeholders in relation to standardisation.
The concept of standardisation is a flexible term when applied to IT innovation. While strictly managed standards do exist (such as those managed by the International Organisation for Standardisation (ISO), the International Telecommunications Union (ITU) and the International Electrotechnical Commission (IEC)) a wide variety of other de facto initiatives, metrics and frameworks that may also be classified as (non strictly) standards.
Sustainability standards, metrics, frameworks and initiatives affect data centres in multiple ways. Some, such as green building certifications, allow data centres to voluntarily demonstrate a certain level of environmental performance. Other standards, such as American Society of Heating Refrigeration and Air-conditioning (ASHRAE)’s temperature standards, can carry the force of de-facto mandates, or form the basis for true mandates later. Sustainability standards can affect how strongly data centres demand efficiency technologies, and sometimes even which technologies they choose.
There are a number of standard organisation sub-groups/committees, which are specifically focused on developing relevant data centre related standards. The project consortium plans to establish contact with these groups (either directly or through intermediaries) and contribute to over the course of the project. For example, it is expected that the Power Usage Effectiveness (PUE) metric will become an official ISO standard by 2014. The CoolEmAll consortium is working with organisations such as the Green Grid to understand how the change in status of PUE will impact its continued adoption by the datacentre industry. Attempts will also be made, where possible, to make some of the outcomes of the project available under open source licenses.
As well as standardization, another fundamental part of the future development strategy for CoolEmAll is to analyse, and engage with potential users, and technology partners. CoolEmAll is developing technologies that could be of use to data centre planners and operators but also licensed to data centre technology suppliers who in turn could sell/license the technology to their customers.
It is expected that some of the potential users and partners the project engages with may eventually transition to become early adopters if they actually deploy the project’s prototype technologies or if they show interest in licensing components etc.
The aim of this engagement process is to gather feedback from these potential users and partners on the technical and strategic direction
of the project. Such input is vital to ensure that the outcomes
of the project are useful to the relevant market.

Potential users and partners will be able to provide real-world testing of the technologies, which result from the project (SVD Toolkit, DEBBs, metrics). Potential partners could also play a vital role in shaping the future development of the project. For example, if an existing technology supplier showed interest in the project, it is possible that the project outcomes could be adjusted to complement the technology already provided by that suppliers
As well as the potential early adopters and partners, another source of external feedback on the project is the project Advisory Board. This group (made up of scientific and commercial experts) would provide an adjunct to the project executive board. The advisory board would provide regular and in-depth feedback on the project’s technical and commercial development.
This kind of cross industry cooperation is fundamental to the European Commission’s vision for projects such as CoolEmAll. The project is ambitious but has a limited time frame. But by collaborating with external organisations, some or all of the technologies developed within CoolEmAll should live on and help bring the concept of holistic datacenter management a little closer to becoming reality.