Monthly bulletin of the IEEE Computer Society Special Technical Community on Sustainable Computing Providing quick access to timely information on sustainable computing. Sponsored by:   Report from the Chairby Anirban Mahanti, NICTA  Martin guided STC-SC through the first year, helping us grow our membership from 24 in January 2012 to 400 by December 2012. Martin pushed the STC-SC officers to work towards creating value for our members. The Register is a very good example of value creation. Other initiatives such as co-sponsorship of conferences, awards for students, and member-to-member engagement are at various stages of development. On behalf of the STC-SC community, I would like to thank Martin for his excellent service. I would also like to thank Giuliano Casale for his service to the STC-SC as secretary/treasurer. The secretary ensures good “governance” by chasing officers for monthly updates and compiling the updates into a coherent report. Chasing officers for their reports can at times be tedious! I want to thank Giuliano for keeping us on track and doing so without ever getting frustrated. Martin hijacked Giuliano for the secretary/treasurer role within the committee that oversees the STC initiative. Danilo Ardagna has taken up Giuliano’s role in the STC-SC. Our STC is well positioned to become even stronger in 2013. I would like to thank everyone for their continuing support and I‘m looking forward to working with the community to make 2013 successful. - Anirban Call for Nominations: Pick of the MonthLast year, the register instituted the Pick of the Month series, a feature article that highlights top research papers and projects within our community. The Pick of the Month was one of the Register's most visible feature articles. This year, we will build upon our success with a slightly modified nomination process. Our goal is to make the process even more democratic. Now, anyone can nominate a paper by simply sending an email to either Prof. Christopher Stewart or Prof. David Chiu. Then vote and converse about your nomination online by: 1. Visiting the URL below and clicking through to the Pick of the Month survey 2. Posting (anonymous) comments describing why the paper deserves pick of the month recognition. You can even comment about why one nominee is better than another. 3. Every month, a nominee with at least 6 votes will be eligible for Pick of the Month. SC STC Information Officers will decide among eligible candidates. Try it out now. http://www.cse.ohio-state.edu/~cstewart/potm In this feature we ask a prominent researcher in the field of sustainable computing to share their journey and lessons along the way with the broader community. In this issue we have the privilege to sit down with Professor Margaret Martonosi from Princeton University, who shares with us some of her interesting research experiences on energy-efficient and high-performance computing and her views looking forward in the sustainable computing field.  Hugh Trumbull Adams ’35 Professor Department of Computer Science Princeton University Currently working on: I work on a range of topics related to high-performance and energy-efficient computing. Currently this includes power-performance tradeoffs in systems ranging from smartphones to chip multiprocessors to large-scale data centers. A particular focus has been on characterizing and exploiting heterogeneous parallelism, where different specialized compute units can be harnessed to retain performance while reducing power. Favorite memory as a student/advisor/researcher: As a professor, my favorite moments are always about watching my undergraduate and graduate students mature in their research abilities and move into their own research careers. It is incredibly rewarding to watch the progression from a student’s first work on experimental research, towards the point where they are creative and independent researchers attacking interesting challenges of their own devising. Could you share a research contribution from your research, and explain why this is something that you are particularly proud? I have two projects of which I am particularly proud. The first of these is Wattch, an architecture-level power modeling tool. When we began work on Wattch, power modeling and power efficiency issues were only slightly a topic of interest for architects; they were typically viewed as being the domain of circuits and devices researchers. Our work demonstrated that early-stage architecture power models could be accurate enough to enable power-aware techniques much earlier in the design pipeline, where they can have more leverage. The second project of which I am particularly proud is ZebraNet. In the ZebraNet project, our group built GPS-based mobile sensor nodes for wildlife tracking, and we used two generations of our designs on zebras in Kenya as part of two major deployment trips. From a biological perspective, ZebraNet afforded biologists with much more detailed and ongoing views of wildlife behavior. From a computing perspective, ZebraNet developed a range of important results: power-efficient peer-to-peer communication protocols, energy-efficient data compression, and several interesting design techniques for sensor systems. ZebraNet demonstrated the utility and value of mobile sensor nodes, of sensor nodes much more capable and coarse-grained than the small and static “smart dust” approach that had prevailed before then. What do you think is (are) the important problem(s) to be solved in the next 10 years within sustainable computing? Energy-efficiency and sustainability issues in computing are at an interesting crossroads. On one hand, computer systems and information technology use much less energy than other economic sectors such as transportation. As a result, our community should be thinking about ways to increasingly leverage more and more computing technology in order to save power elsewhere. This might include improving tele-presence and teleconferencing systems in order to reduce the need for (carbon-intensive) travel to face-to-face meetings. Computer systems can also improve product logistics and e-commerce to achieve sustainability benefits. On the other hand, within the field of sustainable computing, we also need to look inward and improve the energy-efficiency, carbon footprint, and overall sustainability of the computer systems we build. Continuing to improve energy efficiency remains important, particularly at the level of data centers. Some other interesting opportunities here might lie in improving recycling and reuse within the supply chain of our computer systems---true sustainability will require us to reduce the amount of computing equipment that ends up in landfills after relatively few years of use. Do you see the widespread adoption of mobile computing devices playing a central role in sustainability-related research? Around the world, smart phones and mobile handheld devices have increasingly become the dominant computing platform. This brings up interesting research opportunities for location-aware services to improve sustainability. For example, some of our research on SignalGuru has explored lightweight cellphone-based methods for analyzing traffic flow and thereby improving vehicle fuel efficiency by reducing time spent in stop-and-go traffic. What courses and skills are most important for students wanting to work in this area? Making computer systems truly power-efficient requires holistic approaches that span many system layers. Therefore, students will be best prepared if they have taken key courses at different parts of the systems stack. Compilers, Operating Systems and Computer Architectures are particularly important. These prepare students to think creatively about energy-efficient techniques that span the hardware-software interface and that leverage the capabilities of different system layers.
|
Newsletter >
