End-to-end Mapping in Heterogeneous Systems Using Graph Representation Learning
This work addresses the challenge of autonomous programming and optimization in heterogeneous computing systems, offering a domain-specific solution with incremental improvements over existing techniques.
The paper tackles the problem of optimizing code execution on heterogeneous hardware platforms by proposing a programmable graph representation learning framework that predicts the best processor assignments for code segments, achieving a maximum speedup of 6.42x compared to thread-based execution and 2.02x compared to state-of-the-art methods.
To enable heterogeneous computing systems with autonomous programming and optimization capabilities, we propose a unified, end-to-end, programmable graph representation learning (PGL) framework that is capable of mining the complexity of high-level programs down to the universal intermediate representation, extracting the specific computational patterns and predicting which code segments would run best on a specific core in heterogeneous hardware platforms. The proposed framework extracts multi-fractal topological features from code graphs, utilizes graph autoencoders to learn how to partition the graph into computational kernels, and exploits graph neural networks (GNN) to predict the correct assignment to a processor type. In the evaluation, we validate the PGL framework and demonstrate a maximum speedup of 6.42x compared to the thread-based execution, and 2.02x compared to the state-of-the-art technique.