graduation at Engenharia Elétrica from Universidade Federal de Pernambuco (1984), master's at Electric Engineering from Universidade Estadual de Campinas (1988), master's at Electrical Engineering from Princeton University (1993) and doctorate at Electrical Engineering from Princeton University (1997). Has experience in Computer Science, focusing on Computer Systems Architecture, acting on the following subjects: arquitetura de computadores, compiladores, compressão de código, arquiteturas reconfiguráveis and alocação de registradores. (Source: Lattes Curriculum)
| News published in Agência FAPESP Newsletter about the researcher |
| TITULO |
| Articles published in other media outlets (0 total): |
| More itemsLess items |
| VEICULO: TITULO (DATA) |
| VEICULO: TITULO (DATA) |
In this project we join efforts and expertise from several Computer Science re-searchers on High Performance Computing (HPC). Our main goal is to promote high impact research on two areas of HPC: resource management and on new computer architectures. Each area is divided in various topics. Each topic is lead by an experienced researcher and it will be tackled with the help of collaborator…
We propose the creation of a world-class multidisciplinary center dedicated to the development and application of advanced computational modeling techniques for the purpose of meeting the challenges in frontier problems in computational engineering and sciences, and promote substantial advancements in technological innovation, education and dissemination of knowledge in the broader area o…
The increasing demand for online services has led to unprecedentedgrowth in the size and complexity of commercial datacenters.Implementing efficient coordination middleware has then become crucialto ensure consistency and high-availability for the very large volumeof data stored at today's datacenters. In this project we addressdistributed lock manager services, a …
This proposal describes a collaboration project between the Compilers and Computer Architecture teams at University of Alberta and University of Campinas. Both teams have a long history in developing relevant research in the areas covered in this proposal, and will take benefit from this partnership, as detailed below. The main idea is to combine competences and experiences related to par…
(Only some records are available in English at this moment)
Deep learning (DL) methods, especially Convolutional Neural Networks (CNNs), have brought revolutionary advances to many research areas due to their capacity to learn from raw data. However, such astonishing results were obtained at the cost of big and expensive models that run on top of unrealistic computational systems. That kind of huge models, like Stable Diffusion and GPT-3, are norm…
Heterogeneous CPU-FPGA computing platforms, combined with high-level synthesis, form a powerful tool to open up hardware acceleration benefits to problems in Science and Engineering. Examples of problems that can be solved using FPGA acceleration are Gene Regulatory Networks (GRNs), molecular dynamics simulation, seismic imaging, and engineering simulation based on finite difference/eleme…
(Only some records are available in English at this moment)
The task-parallel programming paradigm is more robust than its data-parallel counterpart because it permits the programmer to express himself more freely. However, using task parallelism incurs new challenges of managing dependencies between tasks, synchronization and scheduling. For this reason, in this work we look to identify, characterize and optimize bottlenecks in the LLVM´s impleme…
Contemporary compilers implement various parallelization techniques, however, these techniques are usually restricted to loops that do not have dependencies and are simple to solve, such as reductions of arithmetic operations. These restrictions prevent the parallelization of loops that have dependencies. Loops with loop carry dependencies can present two scenarios: (i) the compiler ident…
The goal of this project is the creation of a compiler-generated parallel code runtime model for remote, distributed execution in computer clusters of the cloud. The ideas here discussed aim at implementing and consolidating OpenMR, a novel programming model, proposed at IC-Unicamp, that extends the existing Open Multi Processing (OpenMP) Application Program Interface (API), to allow a pr…
(Only some records are available in English at this moment)
The study of an organism's metabolism provides information relevant to energy production and phenotype. Constraint-based modeling (CBM) is an approach widely used to study biochemical networks, in particular the reconstruction of the metabolic network through genome-scale metabolic models (GSMM), which contains the reactions, genes that encode each enzyme and its metabolites in an organis…
The staggering speed in which data has been created in the past decade brought a pressing need to efficiently store, and more importantly, extract valuable and accurate information from it. Machine Learning (ML) took the lead of the former due to the great speedup results obtained with General-Purpose Graphics Processors. However, some application domains still lack good ML solutions and …
In this project, we aim to tackle the complexities of programming modern cloud clusters containing heterogeneous components such as CPUs and GPUs. Indeed, modern scientific applications need to exploit all type of parallel programming pattern to reach the performance goals. Computation offloading is a programming model in which program fragments (e.g. hot loops) are annotated so that thei…
(Only some records are available in English at this moment)
| 2 / 2 | Ongoing research grants |
| 7 / 5 | Completed research grants |
| 6 / 6 | Ongoing scholarships in Brazil |
| 29 / 15 | Completed scholarships in Brazil |
| 6 / 6 | Completed scholarships abroad |
| 50 / 34 | All research grants and scholarships |
| Associated processes | |
CV Lattes
ORCID