Publications

12 Results
Skip to search filters

Using MLIR Framework for Codesign of ML Architectures Algorithms and Simulation Tools

Lewis, Cannada L.; Hughes, Clayton H.; Hammond, Simon D.; Rajamanickam, Sivasankaran R.

MLIR (Multi-Level Intermediate Representation), is an extensible compiler framework that supports high-level data structures and operation constructs. These higher-level code representations are particularly applicable to the artificial intelligence and machine learning (AI/ML) domain, allowing developers to more easily support upcoming heterogeneous AI/ML accelerators and develop flexible domain specific compilers/frameworks with higher-level intermediate representations (IRs) and advanced compiler optimizations. The result of using MLIR within the LLVM compiler framework is expected to yield significant improvement in the quality of generated machine code, which in turn will result in improved performance and hardware efficiency

More Details

Low-Communication Asynchronous Distributed Generalized Canonical Polyadic Tensor Decomposition

2021 IEEE High Performance Extreme Computing Conference, HPEC 2021

Lewis, Cannada L.; Phipps, Eric T.

In this work, we show that reduced communication algorithms for distributed stochastic gradient descent improve the time per epoch and strong scaling for the Generalized Canonical Polyadic (GCP) tensor decomposition, but with a cost, achieving convergence becomes more difficult. The implementation, based on MPI, shows that while one-sided algorithms offer a path to asynchronous execution, the performance benefits of optimized allreduce are difficult to best.

More Details

Evaluating Trade-offs in Potential Exascale Interconnect Technologies

Hemmert, Karl S.; Bair, Ray B.; Bhatale, Abhinav B.; Groves, Taylor G.; Jain, Nikhil J.; Lewis, Cannada L.; Mubarak, Misbah M.; Pakin, Scott P.; Ross, Robert B.; Wilke, Jeremiah J.

This report details work to study trade-offs in topology and network bandwidth for potential interconnects in the exascale (2021-2022) timeframe. The work was done using multiple interconnect models across two parallel discrete event simulators. Results from each independent simulator are shown and discussed and the areas of agreement and disagreement are explored.

More Details
12 Results
12 Results