34 citations · 46 across the 7 of their papers we have counts for
7 papers
Memory Allocation for Constant-Bounded Programs
Vinícius Silva, Kael Soares, Márcio Costa e Fernando Magno Quintão Pereira
This work studies memory allocation for constant-bounded programs, whose execution length is syntactically limited for all inputs. Examples of such programs include verified kernel…
Using MLIR Transform to Design Sliced Convolution Algorithm
Victor Ferrari, Marcio Pereira, Lucas Alvarenga +2
This paper proposes SConvTransform, a Transform dialect extension that provides operations for optimizing 2D convolutions in MLIR. Its main operation, SConvOp, lowers Linalg convol…
ConvBench: A Comprehensive Benchmark for 2D Convolution Primitive Evaluation
Lucas Alvarenga, Victor Ferrari, Rafael Souza +2
Convolution is a compute-intensive operation placed at the heart of Convolution Neural Networks (CNNs). It has led to the development of many high-performance algorithms, such as I…
Tensor Slicing and Optimization for Multicore NPUs
Rafael Sousa, Marcio Pereira, Yongin Kwon +5
Although code generation for Convolution Neural Network (CNN) models has been extensively studied, performing efficient data slicing and parallelization for highly-constrai\-ned Mu…
Advancing Direct Convolution using Convolution Slicing Optimization and ISA Extensions
Victor Ferrari, Rafael Sousa, Marcio Pereira +4
Convolution is one of the most computationally intensive operations that must be performed for machine-learning model inference. A traditional approach to compute convolutions is k…
The OpenMP Cluster Programming Model
Hervé Yviquel, Marcio Pereira, Emílio Francesquini +9
Despite the various research initiatives and proposed programming models, efficient solutions for parallel programming in HPC clusters still rely on a complex combination of differ…
Enabling OpenMP Task Parallelism on Multi-FPGAs
R. Nepomuceno, R. Sterle, G. Valarini +3
FPGA-based hardware accelerators have received increasing attention mainly due to their ability to accelerate deep pipelined applications, thus resulting in higher computational pe…