Women in Compilers and Tools June Meetup

In this Meetup, we will be hearing from PhD Candidate at Federal University of Minas Gerais and research intern at Microsoft Research, Angélica Moreira with Static profiling: why should you try it?

In this talk, She will explain the benefits of adopting software-based static profiling, give some useful scenarios where we have tested it and give some insights on possible other useful paths to use it. At the end of this talk, you will be equipped with some concepts like software-based static prediction using ml and heuristics, the state of the art and what compilers currently adopt in this area, you will also get to know the VESPA project. VESPA is a static branch predictor based on machine learning built on top of BOLT. An evaluation of BOLT powered by VESPA on four large benchmarks (clang, GCC, MySQL and PostgreSQL) yields binaries that are 5.47 % faster than the executables produced by clang -O3. You can read the paper VESPA: static profiling for binary optimisation at https://dl.acm.org/doi/abs/10.1145/3485521, to know more details about the VESPA project.

About Angélica:
Angélica Moreira is a 4th-year PhD candidate in the Computer Science Department at the Federal University of Minas Gerais (UFMG) and a research intern at Microsoft Research. During her PhD project, she has been awarded the Microsoft Research PhD Fellowship 2021-2022, the Facebook Emerging Scholar Award 2019-2021 and the 3rd place at IEEE/ACM SRC CGO-2021. She is a member of the Compilers Laboratory (LAC), where she is advised by Professor Fernando Magno Quintão Pereira and Co-advised by Guilherme Ottoni from Meta (previously known as Facebook). She holds an M.Sc degree in Computer Science from the Federal University of Ouro Preto (UFOP) and a B.Sc degree in Computer Science from the Pontifical Catholic University of Minas Gerais (PUC Minas). Angélica’s research interest lies in pursuing the design and implementation of techniques that reduce program binary size and make them run faster on heterogeneous architectures. Her research focuses on combining machine learning and compilation techniques to achieve her goal. Her dream is to contribute to a better world through the development of science and technology.