Hyrax Project [DiCE - Distributed Computing at the Edge] (FCT NOVA/NOVA-LINCS - between February 2017 and January 2018)

A 3rd year's scientific research and development project, of the BSc. degree made in FCT NOVA, done between February 2017 and January 2018. This project it's called Hyrax Project (Crowd-Sourcing Mobile Devices to Develop Edge Clouds). This project it's based in the context of Edge Computing. I did this scientific research and development project, representing NOVA-LINCS (Departamento de Informática da Nova's Labs) and under the supervision of Prof. Hervé Paulino.

• This project, had also the collaboration of Faculdade de Ciências da Universidade do Porto (INESC TEC/CRACS - Centro de Sistemas de Computação Avançada) and Carnegie Mellon University's School of Computer Science;

• Websites/Links:

  • 1) Project's Website:

    • https://hyrax.dcc.fc.up.pt/;

  • 2) Universities' Websites:

    • https://www.fct.unl.pt/;

    • https://sigarra.up.pt/fcup/pt/web_page.inicial;

    • https://www.fc.up.pt/fcup/index.php;

    • https://www.cmu.edu/;

  • 3) Universities' Laboratories' Websites:

    • http://nova-lincs.di.fct.unl.pt/;

    • https://cracs.fc.up.pt/;

    • http://www.cmuportugal.org/;

  • 4) Academic, Industrial and Funding Partners' Websites:

    • http://www.geolink.pt/index.html;

    • http://www.yinzcam.com/;

    • https://www.wavecom.pt/;

    • https://www.it.pt/;

    • https://www.fct.pt/index.phtml.pt;

• I worked in the Services' section of the Hyrax Project, most preciselly, developing a data structure (for DiCE section), to be replicated for Mobile and Handheld Devices, in a notion of collaborative network, using Crowd-Sourcing features and shifting all the data processing and information services closer to the final users;

• Abstract:

  • Edge Computing proposes the shifting of (some) computing tasks from centralized Cloud services to closer to the user, taking advantage of the computing and storage power of personal computers and hand-held devices. Accordingly, computation will be typically performed collaboratively by multiple nodes, which, depending on their current load, may offload or take computation to/from their peers. Moreover, nodes may dynamically join or leave the network, forcing the system to be resilient to Churn. This paper presents a system for collaborative distributed computing in a network of heterogeneous nodes. The system builds on a replicated, distributed and eventually consistent queue of tasks, that requires little coordination. We evaluate our proposal in the context of mobile computing, demonstrating that our solution scales linearly and tolerates Churn efficiently;

• Keywords:

  • Edge/Fog Computing, Distributed Computing, Mobile Computing, Network Computation of Mobile Devices at the Edge, Mobile Devices Replication, Replicate Data Structures, Eventual Consistency;

• Documents of NOVA-LINCS' Hyrax Project (DiCE - Distributive Collaborative Computing at the Edge), in 2018:

  • LaTeX Article/Paper (Portuguese Version);

  • Final Presentation;