Posters

A global Personal Energy Meter

Simon Hay, Andrew Rice and Andy Hopper

Every day each of us consumes a significant amount of energy either directly through transportation, heating or use of appliances or indirectly from our needs for production of food, manufacture of goods or provision of services. We envisage a Personal Energy Meter (PEM) which can record and apportion an individual's energy usage in order to provide baseline information and incentives for reducing the environmental impact of our lives.

Presented at: Ubiquitous Computing at a Crossroads: Art, Science, Politics and Design (UbiComp Grand Challenge).

Powering data centres with surplus renewable energy

Sherif Akoush, Andrew Rice and Andy Hopper

Generation capacity from renewable energy sources varies according to environmental conditions. By moving computing jobs around the globe, we can make use of surplus energy which could otherwise be lost. We consider wind and solar power worldwide to build a simulator. This simulator includes a model for internet as communication infrastructure to move data. We design load optimizer (balancer/un-balancer) that takes into account power fluctuations, different workload characteristics and SLA guarantees.

Computing for the future of the planet

Andy Hopper, Andrew Rice, Alastair Beresford and Robert Harle

Computing has had a huge impact on the way we work and spend our leisure time. In the last 10 years alone, the Internet has revolutionised communications and minaturisation has put immense computing power literally in the palm of our hands. Imagine what could be achieved if the power of computing was truly focussed and harnessed to deliver environmental benefits.

Computing for the future of the planet

Andrew Rice, Andy Hopper, David Cottingham, Jonathan Davies,Mbou Eyole-Monono, Anthony Hylick and Brian Jones

There are many issues threatening the future of our planet. The impact of climate change is a prominent topic. However, there are numerous other problems such as overfishing, desertification, deforestation and rising energy demands. These are all compounded by a global population which is expected to grow to 9 billion people by 2050. We believe that computing has a significant role to play in solving these problems and have highlighted four initial areas of research.

Hard drive power consumption uncovered

Anthony Hylick, Andrew Rice, Brian Jones and Ripduman Sohan

Attempts to reduce power consumption have mainly focused on maximizing standby periods. This approach is based on the assumption that the mechanics of a drive dominate the electronics in terms of power consumed. Our fine-grained measurements indicate a more complex situation. We present results which show the importance of a detailed understanding of power consumption and identify the need for a more expressive API between the OS and hardware devices to maximize power efficiency.

PASv: Power-aware servers through virtualization

Anthony Hylick, Andrew Rice, Brian Jones and Ripduman Sohan

By providing a platform to maximize utilization and efficiency, virtualization has emerged as a key solution in helping to alleviate the power consumption crunch experienced by data centers of all capacities. We aim to collect detailed, online power measurements of both hardware and software.

Pedal Powered Computing

Brian Jones and Andrew Rice

The cost, both environmental and economic, of electricity generation is significant. Current initiatives in sustainable power generation make use of tidal, wind and solar power. This work seeks to investigate the possibility of people directly generating their own electricity.

Cantag

Andrew C. Rice, Alastair R. Beresford and Robert K. Harle

Cantag is a marker-based machine vision system designed to allow the user to select, and compare, different tag designs and tracking algorithms. It allows investigation into fundamental properties and limitations of particular 2-dimensional marker tag designs. Cantag is open source software written in C++.

Demo: investigating algorithms in the machine vision pipeiline

Andrew C. Rice, Alastair R. Beresford and Robert K. Harle

Cantag is a marker-based machine vision system designed to allow the user to select, and compare, different tag designs and tracking algorithms. It allows investigation into fundamental properties and limitations of particular 2-dimensional marker tag designs. This demonstration shows a number of possible techniques for recognising square tags. The image processing pipeline may be reconfigured dynamically by displaying the relevant marker tag to the camera.