Talking About Green

The impact of Climate Change on Academic Research

Cyber-Infrastructure is part of the problem, but it is also part of thesolution.

NSF GreenLight at CAL-IT2 and PROMPT G-NGI initiatives to reduce GreenhouseGas (GHG) impact of academic research.

To date most academic researchers have not been particularly concerned aboutthe impact of climate change on their academic research.  To manyresearchers climate change only affects big polluters such as coal plantsand owners of SUVs. Surprisingly few members of the research communityappreciate the dramatic changes that will be required in the next couple ofyears, if we hope to slow down the rate of temperature increase in the nextdecade (never mind trying to stop or reverse climate change as result of GHGemission).  Every aspect of our lives will be fundamentally altered associety starts to recognize the severity of the problem, including, andespecially in the way we carry out academic research.

Governments around the world are already starting to impose carbonneutrality on public sector institutions such as universities, schools andhospitals. This strategy is becoming increasing popular with governments asthe public sector is a large part of the economy and therefore a majorcontributor to GHG emissions. More importantly it avoids the anguish andcontroversy of imposing carbon taxes on the voting public.

University researchers and funding agencies had better be prepared for thesedevelopments. The concept of mandated carbon neutrality will spread likewildfire once governments around the world discover its many benefits.

Fortunately the academic research community already has many the tools athand, not only to be carbon neutral, but perhaps even achieve zero carbonsustainability.  It is becoming evident that one of the most importantscientific tools for research exploration at our universities iscyber-infrastructure. Through the use of networks, grids, virtualization andremote instrumentation and laboratories it is the one research tool that canhelp reduce GHG emissions at our campus.

But currently cyber-infrastructure located "on campus" is part of theproblem. At many of our universities it is increasingly a major source ofGHG in its own right through the power it consumes. The beauty and power ofcyber-infrastructure is it removes the restriction that physical facilitiesneed to be located on campus. With high speed optical networks these samefacilities can be located at zero carbon data centers anywhere in thecountry that have easy access to renewable energy. Relocatingcyber-infrastructure to renewable energy sites will be much cheaper thantrying to purchase renewable power locally on campuses in our cities, as theuniversity will be competing with businesses for that same power.

Two important initiatives are now underway which will help academicresearchers address the challenge of reducing their carbon footprint throughthe use of cyber-infrastructure. The first is the PROMPT program for NextGeneration Internet to Reduce Global Warming. This is an internationalpartnership being led by PROMPT in Montreal with partners from the worldincluding Australia, The Netherlands, United States, China etc. It is aresearch and commercialization initiative to help carry out research andcommercialize the next generation Internet technologies being developed atour universities such as wireless devices, sensors, instruments and networksthrough the use of virtualization and SOA, etc.  The initiative is unique inthat rather than negotiating traditional licenses and royalties, paymentsfor companies who adopt the technology will be made through the purchase ofcarbon credits. PROMPT will also work with universities in helping themdevelop research practices and procedures in order to reduce their carbonfootprint.  PROMPT proposes to develop a set of testbeds in Canada and withits international partners to develop the necessary protocols to test verifyand audit the actual carbon credits in compliance with ISO 14064 that willbe possible through the application of next generation Internettechnologies.

The second important initiative which is also important to the PROMPTprogram is to actually measure the energy savings and CO2 reductions ofcyber-infrastructure equipment.  To this end the NSF has awarded a researchteam at CAL-IT2 funding for a project called GreenLight. This project,measures, monitors, and optimizes the energy consumption of large-scalescientific applications from many different areas. The work enablesinter-disciplinary researchers to understand how to make "green" (i.e.,energy efficient) decision for IT computation and storage, thus helping tore-define fundamentals of systems engineering for a transformative concept,that of green CyberInfrastructure .

Academic research is about to go through a major revolution in the way andhow it is carried out.  Cyber-infrastructure will play a critical role.Researchers and institutions that are the first to adopt to these new way ofdoing research will be the big winners in the future. --BSA]

For additional information:

The Impact of Climate Change on Academic Researchhttp://www.slideshare.net/bstarn/impact-of-climate-change-on-academic-research/

PROMPTwww.promptinc.org

Greenlighthttp://nsf.gov/awardsearch/showAward.do?AwardNumber=0821155

This project, developing an instrument called GreenLight, measures,monitors, and optimizes the energy consumption of large-scale scientificapplications from many different areas. The work enables inter-disciplinaryresearchers to understand how to make ?green? (i.e., energy efficient)decision for IT computation and storage. Consequently, an experienced teammight be able to make deep and quantitative explorations in advancedarchitecture, including alternative circuit fabrics such as FieldProgrammable Gate Arrays (FPGAs), direct-graph execution machines, graphicsprocessors, solid-state disks, and photonic networking. The enabledcomputing and systems research will yield new quantitative data to supportengineering judgments on comparative ?computational work per watt? acrossfull-scale applications running at-scale computing platforms, thus helpingto re-define fundamentals of systems engineering for a transformativeconcept, that of green CyberInfrastructure (CI). Keeping in mind that the ITindustry consumes as much energy (same carbon footprint) as the airlineindustry, this project enables five communities of application scientists,drawn from metagenomics, ocean observing, microscopy, bioinformatics, andthe digital media, to understand how to measure and then minimize energyconsumption, to make use of novel energy/cooling sources, and employmiddleware that automates optimal choice of compute/power strategies. Theresearch issues addressed include studying the dynamic migration ofapplications to virtual machines for power consumption reduction, studyingthe migrations of virtual machines to physical machines to achieve networklocality, developing new power/thermal management policies (closed loop,using feedback from sensors), classifying scientific algorithms in thecontext of co-processing hardware such as GPUs and FPGAs, and developingalgorithms for resource sharing/scheduling in heterogeneous platforms. Thefull-scale virtualized device, the GreenLight Instrument, will be developedto measure, monitor, and make publicly available (via service orientedarchitecture methodology), real-time sensor outputs, empowering researchersanywhere to study the energy cost of at-scale scientific computing. Hence,this work empowers domain application researchers to continue to exploitexponential improvements in silicon technology, and to compete globally.Although the IT industry has begun to develop strategies for ?greening?traditional data centers, the physical reality of modern campus CI currentlyinvolves a complex network of ad hoc and suboptimal energy environments indepartmental facilities. The number of these facilities increases extremelyfast creating campus-wide crisis of space, power, and cooling due to thevalue of computational and data intensive approaches to research. Thisproject addresses these important issues offering the possibility toimprove. Broader Impacts: The project enables researchers to carry-outquantitative explorations into energy efficient CyberInfrastructure (CI) andto train the next generation of energy-aware scientists. It enlists graduatestudents from five disciplinary projects, involves minority servinginstitutions, and is likely to have direct impact on commercial componentsof the nation?s CI.