Deepframe: A Profile-Driven Compiler for Spatial Hardware Accelerators

Guha, Anirban; Vedula, Naveen; Shriraman, Arrvindh

doi:10.1109/pact.2019.00014

Cited by 6 publications

(1 citation statement)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cavazos et al [10] use profile data as input features to a regression model that predicts the best compiler flags. DeepFrame [18] incorporates deep learning methods to learn the most likely paths during execution and offload the regions to FPGAs. Though profiler-guided optimizations can automatically adjust code based on rules or models, they only cover a subset of all the available optimizations.…”

Section: Related Workmentioning

confidence: 99%

GPA: A GPU Performance Advisor Based on Instruction Sampling

Zhou

Meng

Sai

et al. 2020

Preprint

View full text Add to dashboard Cite

Developing efficient GPU kernels can be difficult because of the complexity of GPU architectures and programming models. Existing performance tools only provide coarse-grained suggestions at the kernel level, if any. In this paper, we describe GPA, a performance advisor for NVIDIA GPUs that suggests potential code optimization opportunities at a hierarchy of levels, including individual lines, loops, and functions. To relieve users of the burden of interpreting performance counters and analyzing bottlenecks, GPA uses data flow analysis to approximately attribute measured instruction stalls to their root causes and uses information about a program's structure and the GPU to match inefficiency patterns with suggestions for optimization. To quantify each suggestion's potential benefits, we developed PC samplingbased performance models to estimate its speedup. Our experiments with benchmarks and applications show that GPA provides an insightful report to guide performance optimization. Using GPA, we obtained speedups on a Volta V100 GPU ranging from 1.03× to 3.86×, with a geometric mean of 1.22×.

show abstract

Section: Related Workmentioning

confidence: 99%