GMAP researcher wins 2nd place in WSCAD 2022 Thesis and Dissertation Competition

Adriano Garcia

1 year ago

The Computer and High-Performance Computing Thesis and Dissertation Competition (WSCAD-CTD) occurred between October 19th and 21st. The competition was held with the XXII Symposium on High-Performance Computing Systems (WSCAD 2022) in Florianópolis city. Master’s and Ph.D. theses defended from July 1st, 2021, in Computer Architecture and High-Performance Computing were accepted for submission.

GMAP researcher Adriano Vogel presented his Ph.D. thesis entitled “Self-adaptive abstractions for efficient high-level parallel computing in multi-cores”. The thesis was advised by Dr. Luiz Gustavo Fernandes from PUCRS and Prof. Dr. Marco Danelutto from UNIPI. In addition, the thesis was co-advised by Prof. Dr. Dalvan Griebler from PUCRS. The researcher had already been approved with honor in defense of his Ph.D. thesis in March of this year.

This work aimed to provide self-adaptive abstractions, in which self-adaptation transparently manages the executions while the parallel programs are running (at run-time). The main goals of this work are to increase the adaptation space to be more representative of real-world applications and make self-adaptation more efficient with comprehensive evaluation methodologies, which can provide use cases demonstrating the true potential of self-adaptation. The doctoral dissertation provided the following scientific contributions:

I) Categorizations, a taxonomy, a catalog of self-adaptation optimization, and a discussion of research challenges and perspectives of self-adaptive executions in parallel stream processing;

II) A conceptual framework for decision-making in self-adaptive parallel executions;

III) Strategies for a self-adaptive number of replicas for applications with a single parallel stage. The strategies support non-functional requirements such as latency and throughput and SLO for managing resources utilization, minimizing self-adaptation overhead, and providing seamless parallelism management;

IV) Mechanism, models, and strategies for self-adaptive applications’ composition structures.

V) Mechanism and strategy for a self-adaptive number of replicas in complex compositions within the applications’ composition structures.

More information about the WSCAD-CTD 2022 can be found on the official website.