Data Centers and HPC: the Energy Challenge

If you were asked to list industries whose carbon footprint contributes dramatically to global warming and represents a threat to our planet’s delicate climate balance, you would probably not think of IT as one of the top “energy hogs”.

You would be wrong. IT is rapidly climbing in this not-so-virtuous ranking, with its double digit growth rate in energy consumption dwarfing the transportation industry’s own of 1% per year. According to the industry consortium GreenTouch, the IT industry accounts today for roughly 2% of the world’s total energy consumption, comparable to the airline industry.

While the fantastic computing power that technology is making available to businesses and to strategic fields such as science, applied research and industrial R&D is enabling progress, ultimately contributing to the better good of human kind in ways rarely seen before, the desire for even faster progress is boosting the need for ever increasing computing power and for access to huge amounts of data, all but accelerating this segment’s demand for energy.

Data centers are responsible for a large chunk of the energy consumed by IT. According to the U.S. Environmental Protection Agency (EPA) the amount of energy consumed by data centers doubled between 2000 and 2006 alone. This trend slowed down in 2007 amid economic crisis and better data centers efficiency, to start accelerating again in the last 2 years. A recent report from the Natural Resources Defense Council (NRDC) claims waste and inefficiency in U.S. data centers – that consumed a massive 91 bn kWh of electricity in 2013 – will increase to 140 bn kWh by 2020, the equivalent of 50 large (500 megawatt) power plants.

This has become largely unsustainable.

In an ideal situation, IT equipment should use 100% of the energy consumed by a data center. Unfortunately, reality is different. The percentage of energy used for IT equipment varies between 60% and 30% of the total energy consumed by the whole data center. The parameter that best measures data center energy efficiency is PUE (power usage effectiveness). The closer PUE is to 1 the better: a PUE of 2.5 means that for every Watt consumed by the data center 1 is used  for the IT equipment and 1.5 Watt goes for cooling or other not essential activities. A PUE of 1 means that 100% of the energy used by the data center goes into IT equipment.

A 2014 Uptime Institute annual data center survey reveals that data center power usage efficiency (PUE) metrics have plateaued at around 1.7 after several years of steady improvement.

The main reasons of this low efficiency are energy waste in the electric conversion needed to power equipment (transformers, rectifiers, UPS) and energy used to cool IT equipment through chillers and CRAC units.

There are many technologies that are being used to improve energy efficiency in data centers: virtualization, hot and cold aisle containment, increase of thermal envelope, air flow optimization, DCIM (Data Center Infrastructure Management). These are all low hanging fruits that allowed for an increase in efficiency, but they don’t allow reaching sustainable levels.

HPC (High Performance Computing) has long seen energy cost and availability as the biggest challenges for future developments. It is not a surprise that the HPC segment is currently adopting the most advanced solutions for energy efficiency, aiming to reduce consumption of both IT equipment and datacenter infrastructure, as well as reusing the thermal energy servers produce.

Designing more energy efficient systems means taking an approach where efficiency comes first. This implies making HW and SW design choices that maximize performance within a target power budget, leveraging heterogeneous architectures, accelerators, solid state disks, no-fan liquid cooled systems and in general choosing always the components that can guarantee more efficiency.

Wherever possible, the goal must be to achieve “free cooling”, meaning the IT equipment should be cooled without using additional energy, for instance by eliminating chillers, thus pushing down the datacentre PUE to levels around 1.05, very close to the ideal value of 1. Free cooling is only feasible when the coolant has a temperature higher than the external air temperature. If the outside temperature is very low, for instance at high latitudes or elevations, the coolant may be air, in all other cases it has to be liquid, typically water, warm enough to be cooled with outdoor air also in hot seasons. The only way to use warm water to cool IT equipment is to bring it as close as possible to where the heat is generated, at direct contact with the components (“direct liquid cooling”).

The development and optimization of the technologies leading to better energy efficiency in IT, and in HPC, requires non trivial R&D investments by the manufacturers of large scale computers used in datacenters and HPC centers. While better energy efficiency contributes significantly to lowering the Total Cost of Ownership of IT equipment, the R&D costs manufacturers sustain may lead to higher prices for equipment built according to energy efficiency criteria.

As for many other areas around carbon footprint reduction, “doing the right thing” may end up being economically less attractive than doing the “wrong” one.

While in other industrial and domestic segments (transportation, heating, renewable energy generation), policy, recommendations, stricter regulations and incentives start yielding tangible results, the IT industry and HPC have been only marginally touched by such initiatives.

As long as setting up an energy inefficient datacentre is an economically viable option for IT equipment owners, it is unlikely that substantial progress will be made towards reversing a dangerous trend.

While the issue is a planetary one, now is a good time for Europe to take it in its own hands and show the planet the way towards a more responsible and energy conscious future for the IT industry and High Performance Computing.

Data Centers and HPC: the Energy Challenge

How EU-funded research eInfrastructures could address sustainability, innovation and data-related challenges

In a previous post I summarised the main challenges faced by research infrastructures as identified by Horizon 2020 Research Infrastructure Advisory Group. The challenges are sustainability, innovation and data exploitation. In this 14-minute videoblog, I share how I think EU-funded research eInfrastructures could address these challenges.

As usual, my objective is not to be prescriptive, but rather to stimulate debate. For those of you interested in understanding a little bit better how I think about research eInfrastructures, this videoblog might be of interest as well.

If you are in a rush and don’t have time to watch the video, you can download the slides to read them at your leisure.

Please share your comments below and happy watching 🙂

How EU-funded research eInfrastructures could address sustainability, innovation and data-related challenges

European Exascale vision and strategy on Big Data and Extreme Computing

I have been invited to participate in the panel discussion “European Exascale vision and strategy on Big Data and Extreme Computing” organised in the context of Big Data and Extreme-scale Computing for Europe workshop that takes place in Barcelona, Spain on January 28, 2015. I thank the organisers for their kind invitation. This blog post summarises the main points of my intervention.

Applications are a strategic asset for Europe in the HPC arena. Europe is leader in HPC-powered applications that address societal challenges or are important for industrial competitiveness. A considerable demand for HPC is also present in emerging domains such as Big Data, High-Performance Data Analytics, Global System Science, as well as flagship projects such as the Human Brain Project.

However, only very few applications using HPC really take advantage of current petaflop machines. Current application owners and users find it a challenge to adapt to increasing performance and parallelism and to new architectures.

In the previous European research framework programme, FP7, no specific activity was dedicated to HPC application work. A certain amount was carried out inside PRACE, mainly to scale applications for efficient use of Tier-0 machines and to offer some support to industrial applications, but not in a systematic way and not always led by the application users as we believe it should be the case.

Our European HPC strategy identifies “Excellence in applications” as one of the three critical pillars in Horizon 2020, the current European research framework programme. This pillar focuses on supporting the establishment of Centres of Excellence (CoEs) for computing applications.

In our view, these CoEs should further consolidate the EU’s strong position in HPC applications by coordinating and stimulating parallel software code development and scaling, and by ensuring the availability of quality HPC software to users. There is a need to gather multidisciplinary teams, associating computer scientists, mathematicians and scientists to define better computational methods and algorithms

The CoEs will focus and coordinate the application work of HPC in scientific or industrial domains that are most important for Europe. The objective of the CoEs will be to develop a culture of excellence, both scientific and industrial, placing computational science at the centre of scientific discovery and industrial competitiveness

CoEs will aim at the development, provisioning and support of leading edge software and associated expertise and skills for both scientific and industrial applications. Moreover, increased productivity and competitiveness require that codes are well maintained and validated by communities rather than individual users. They should be made available through clear license terms with professional support, rather than the “best effort” approach used for many codes today.

Preparing applications for exascale imply that very advanced software and new computational methods and algorithms must be made available and adopted by practitioners, radically changing the way that large-scale applications are conceived and programmed. We expect that this activity will link with the support to applications/algorithmic developments that are carried out in the FET-HPC programme (also part of Horizon 2020).

Last but not least, CoEs should work in synergy with the other two pillars of the HPC strategy: the pan-European HPC infrastructure, and the technology-supply effort towards exascale. This could be done by identifying suitable applications for co-design activities relevant to the development of HPC technologies towards exascale, and the collaboration with supercomputing centres for the re-use of common tools and methods and the validation the new approaches to applications

The topics of the CoEs will be chosen following the recently closed call for proposals. Our goal is not to cover all possible HPC application areas, but to have at the end of the day some excellent ones that can become Europe and world-wide references in their specific domain

European Exascale vision and strategy on Big Data and Extreme Computing

Key Performance Indicators and cooperation among eInfrastructure projects

We have just launched the grant preparation phase for the first big set of eInfrastructure projects under Horizon 2020 and we have written to the project coordinators to underline the issues they have take into consideration when preparing their technical annexes.

All the issues raised in the letters are consistent with our legal framework, but there are two that I consider really relevant in our current context: Key Performance Indicators (or KPIs) and cooperation across projects.

We have reminded project coordinators about the requirement to monitor the performance of their projects and gather relevant data to help in assessing their impact. To this end, the consortia are asked to establish the most relevant KPIs and monitor them during the project’s duration. These KPIs will be reported periodically to the us.

In addition, we have asked the projects to actively participate in concertation activities, consultations and other meetings and events related to e-infrastructure. The objective is to optimise synergies between projects by providing input and receiving feedback from other stakeholders. This activity should also be described in the technical annex.

In this context, we will group the e-infrastructure projects that successfully complete the grant preparation phase into clusters of related projects. I will post here information on the cluster relevant to all projects at the finalisation of the grant preparatory phase. We expect that project clusters proactively prepare a common presentation for the first concertation meeting we plan to organise in November 2015.

Key Performance Indicators and cooperation among eInfrastructure projects

Our biggest challenges in 2015

Early January is always the time to think about last year’s achievements and what’s ahead of us. And it is always the same: no matter how great the previous year was, the challenges ahead feel overwhelming. This post is about the biggest challenges being faced by the Research Infrastructure (RI) community as identified by the Horizon 2020 Advisory Group on European Research Infrastructures. After having read this post, please tell me if I have reasons to feel overwhelmed. Continue reading “Our biggest challenges in 2015”

Our biggest challenges in 2015

Why is the Juncker 315 billion euro investment plan important for research e-infrastructures?

The 315 billion euro investment package launched by the Juncker Commission was endorsed by the European Council in its meeting of 18 December 2014. In this blog post I will try and illustrate why this plan is important for the Research e-Infrastructures in Europe by briefly describing the plan, explaining the role of the Member States & regions and, finally, providing an extract of the illustrative list of projects relevant to e-infrastructures.

Continue reading “Why is the Juncker 315 billion euro investment plan important for research e-infrastructures?”

Why is the Juncker 315 billion euro investment plan important for research e-infrastructures?

Some reflections after #RDAITEU2014

These are the main points that I take back to Brussels after one day and a half of inspiring presentations and fruitful conversations with members of the RDA and research e-Infrastructures communities.

RDA brings together all stakeholders in the field of research data. The institutional and governmental support is strong and scientific communities are engaged in solving specific data-related problems. The momentum must be sustained perhaps also through financial incentives. Involvement of computer scientists and industry must also be reinforced. Sharing of basic services across scientific communities must be pursued. On this the most advanced communities must take the lead. The initiative launched by Cristina Messa when closing the workshop will help in this respect. Continue reading “Some reflections after #RDAITEU2014”

Some reflections after #RDAITEU2014