Sagitarii: Data Science Workflow System

I am currently working in a distributed computing project called Sagitarii.

Sagitarii is a Data Science Workflow System that uses distributed computing to process scientific workflows. The differential of Sagitarii is the application of the concepts of relational algebra to create efficient operators for the activities. This innovative method can be seen in detail in the article An Algebraic Approach for Data-Centric Scientific Workflows, ( Prof. Eduardo Ogasawara, da Escola de Informática e Computação (EIC) – CEFET-RJ and others ).

 

Workflow Design

Workflow Design

Another great advantage of Sagitarii is the application of the concepts of Distributed Simulation in its communication protocol. Treating the core processors of processing nodes as a Federate, direct communication is possible between the server and the cores, increasing the efficiency of data transfer.

Sagitarii as a Federation

Sagitarii as a Federation

Sagitarii has an efficient and intuitive web interface, making it easy to use and accessible for researchers who have no knowledge in programming languages.

Node Monitor screen

Node Monitor screen

Using this interface, you can monitor the entire progress of the scientific experiments execution, as well as the production of data and their origin with real-time detailed reports.

Sagitarii is still under development and being tested in a controlled environment in CEFET-RJ laboratories. Any questions can be directed to me and I‘ll contact the project manager, Prof. Ogasawara.

Brazilian Portuguese version

Add Comment