Impact case: Sustainable Distributed Systems for Sustainable Research and Education


Summary of the impact
By carefully building architectures, designs, implementations, and experimentation of distributed computer systems that aims at improving the sustainability of our society, the department is providing unique answers to both the public and the industry major concerns in northern Norway. The systems we build and are building are useful to quantify and understand important factors and parameters that impact our everyday life in Norway and northern Norway.

Underpinning Research

A complete example that shows the impact of the department in understanding important factors impacting northern Norway and discovering how to build this specific type of systems is the Distributed Arctic Observatory (DAO project).

COAT, the Climate-ecological Observatory for the Arctic Tundra, aims at giving “a robust documentation and predictions on climate change impacts on biodiversity, natural resources and ecosystem processes in the Norwegian Arctic” (https://coat.no/). However, the context of the Arctic Tundra, being vast, hard to reach, dangerous, extremely limited in infrastructure (both network and energy), make it extremely complicated to have resilient, connected, frugal and autonomous distributed systems observing the Arctic Tundra. As we speak, less than one per cent of such an ecosystem is observed in-situ (i.e. at a very fine grain level, beyond the help of coarse grain observations, like satellites), needed to understand the evolution of specific parameters (e.g, evolution of a certain type of fauna, flora, type of animals, ecological parameters, etc.).

The Distributed Arctic Observatory aims at providing architectures, designs, implementations, experimentation, and prototypes to give insights to the computer science and interdisciplinary research communities on how to build systems deployed and observing such environments.

The DAO project is motivated by many key challenges, highlighted as current and future challenges for “peace and prosperity for people and the planet” by the United Nations (https://sdgs.un.org/goals). By building a distributed observatory for an eco-system like the arctic tundra (i.e. endangered, very sensitive to climate change, very isolated in infrastructure and resources), this project directly contributes to goal 13 “Take urgent action to combat climate change and its impacts”, goal 15 “Protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss” (see publication 1, 2 and 3 below). 

Following the discoveries of the DAO-Cyber Physical System, systems in such environments could cover larger areas, gather, and crunch more data, be more autonomous and more frugal.

The difficulties creating prototypes is a known challenge. Building prototypes for constrained environment, such as the DAO, is an even harder challenge. To simplify and ease such challenges and its dissemination, we created “UiT’s IoT to (extreme) edge testbed”. Such a testbed aims at providing a platform for students, researchers, and industries to provide and simplify the usage and testing of prototypes (both software and hardware) in such constrained environment, with external monitoring for key metrics like energy consumption.

By building prototypes with off the shelf hardware and making them widely available, the DAO project and the “UiT’s IoT to (extreme) edge testbed” contributes to goal 17, where “affordable technological solutions have to be developed and disseminated widely in the next fifteen years” (https://sdgs.un.org/topics/technology).

Details of the impact

The researched underpinned can be directly included in interdisciplinary research observing environments in hard-to-reach scenarios. We document the proposed results about the architectures, designs, implementations and prototypes with reproducible experimentation, methodologies, and results. This is a key asset in creating useful and directly reusable results for other researchers. This documentation is done through multiple papers published in highly recognized and selective peer reviewed international IEEE and ACM conferences, highly ranked in international rankings.

We aim at impacting both researchers in computer science but also others research fields. It can also impact stakeholders, deciding on how and what should be done next in terms of decisions to, for example, save endangered and difficult to observe eco-systems like the Arctic Tundra.

Such impact on many interdisciplinary fields can be shown by looking at the research papers using our research and citing them. A good example is the paper entitled “Leveraging energy-efficient non-lossy compression for data-intensive applications” cited by papers in computer science, geoscience, and environmental research. Another one would be “Trading data size and CNN confidence score for energy efficient cps node communications” cited by papers in system computer science but also industrial informatics. Such wide reachability on recent papers shows a high impact and reusability of our proposed approaches and research.

Participants

The CPS group lead and together with the AGC group participated in the effort of creating, maintaining, and extending the department’s research around the observation of the Arctic Tundra.

Central persons with roles at the department working on the efforts around the DAO project are Profs. Otto Anshus, John Markus Bjørndalen, Loic Guegan, Hoai Phuong Ha and Issam Raïs.

Central persons with roles at the department working on the efforts around the Extreme edge testbed are Profs. Issam Raïs, Loic Guegan, Otto Anshus and John Markus Bjørndalen.

Sources to corroborate the impact

Manish Parashar and Hermann Härtig were invited to participate and contribute to the research around the observation of the Arctic Tundra done by the CPS group. They can be contacted to verify the claims made with regards to this impact case.

MANISH PARASHAR, Scientific Computing and Imaging Institute, 72 S Central Campus Drive, Room 3750, Salt Lake City, UT 84112

Telephone: (801) 585-1867
Email: parashar@sci.utah.edu 
https://www.manishparashar.org/ 

HERMANN HÄRTIG, Andreas-Pfitzmann-Bau, Raum 3088 Nöthnitzer Straße 46, 01187 Dresden, Germany

Telephone +49 351 463-38282
Email: hermann.haertig@​tu-dresden.de
https://tu-dresden.de/ing/informatik/sya/professur-fuer-betriebssysteme/die-professur/haertig?set_language=en

References to the research

Publication 1:
Experiences Building and Deploying Wireless Sensor Nodes for the Arctic Tundra, 21st IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid), 2021,
https://doi.org/10.1109/CCGrid51090.2021.00047
https://ieeexplore.ieee.org/document/9499357

 Publication 2:
Impact of loosely coupled data dissemination policies for resource challenged environments, 22nd IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid), 2022,
https://doi.org/10.1109/CCGrid54584.2022.00062
https://ieeexplore.ieee.org/abstract/document/9826006

Publication 3:
UAVs as a Leverage to Provide Energy and Network for Cyber-Physical Observation Units on the Arctic Tundra, IEEE International Conference on Distributed Computing in Sensor Systems (DCOSS), 2019,
https://doi.org/10.1109/DCOSS.2019.00114
https://ieeexplore.ieee.org/abstract/document/8804827

Publication 4:
Trading Data Size and CNN Confidence Score for Energy Efficient CPS Node Communications, 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid), 2020,
https://doi.org/10.1109/CCGrid49817.2020.00-46
https://ieeexplore.ieee.org/abstract/document/9139716



Members:

Otto Johan Anshus
Loïc Guégan
Issam Rais
John Markus Bjørndalen