Operator Fusion in XLA: Analysis and Evaluation

• URL: http://arxiv.org/abs/2301.13062v1
• Date: Mon, 30 Jan 2023 17:03:33 GMT
• Title: Operator Fusion in XLA: Analysis and Evaluation
• Authors: Daniel Snider, Ruofan Liang
• Abstract summary: XLA is the most common machine learning (ML) compiler. We show how different fusion decisions in XLA are made in practice. We implement several XLA kernel fusion strategies that can achieve up to 10.56x speedup.
• Score: 1.2183405753834562
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Machine learning (ML) compilers are an active area of research because they offer the potential to automatically speedup tensor programs. Kernel fusion is often cited as an important optimization performed by ML compilers. However, there exists a knowledge gap about how XLA, the most common ML compiler, applies this nuanced optimization, what kind of speedup it can afford, and what low-level effects it has on hardware. Our paper aims to bridge this knowledge gap by studying key compiler passes of XLA's source code. Our evaluation on a reinforcement learning environment Cartpole shows how different fusion decisions in XLA are made in practice. Furthermore, we implement several XLA kernel fusion strategies that can achieve up to 10.56x speedup compared to our baseline implementation.

Related papers

• Liger Kernel: Efficient Triton Kernels for LLM Training [6.373771349397682]
  Training Large Language Models (LLMs) efficiently at scale presents a formidable challenge, driven by their ever-increasing computational demands. We introduce Liger- Kernel, an open-sourced set of Triton kernels developed specifically for LLM training. With kernel optimization techniques like kernel operation fusing and input chunking, our kernels achieve on average a 20% increase in training throughput and a 60% reduction in GPU memory usage.
  arXiv  Detail & Related papers  (2024-10-14T18:17:01Z)
• LLM-Aided Compilation for Tensor Accelerators [6.709490736813537]
  We discuss how large language models (LLMs) could be leveraged to build a compiler for hardware accelerators. Specifically, we demonstrate the ability of GPT-4 to achieve high pass rates in translating code to the Gemmini accelerator. We also propose a 2-phase workflow for utilizing LLMs to generate hardware-optimized code.
  arXiv  Detail & Related papers  (2024-08-06T19:10:25Z)
• KGym: A Platform and Dataset to Benchmark Large Language Models on Linux Kernel Crash Resolution [59.20933707301566]
  Large Language Models (LLMs) are consistently improving at increasingly realistic software engineering (SE) tasks. In real-world software stacks, significant SE effort is spent developing foundational system software like the Linux kernel. To evaluate if ML models are useful while developing such large-scale systems-level software, we introduce kGym and kBench.
  arXiv  Detail & Related papers  (2024-07-02T21:44:22Z)
• StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback [58.20547418182074]
  We introduce StepCoder, a novel framework for code generation, consisting of two main components. CCCS addresses the exploration challenge by breaking the long sequences code generation task into a Curriculum of Code Completion Subtasks. FGO only optimize the model by masking the unexecuted code segments to provide Fine-Grained Optimization. Our method improves the ability to explore the output space and outperforms state-of-the-art approaches in corresponding benchmarks.
  arXiv  Detail & Related papers  (2024-02-02T13:14:31Z)
• JaxMARL: Multi-Agent RL Environments and Algorithms in JAX [105.343918678781]
  We present JaxMARL, the first open-source, Python-based library that combines GPU-enabled efficiency with support for a large number of commonly used MARL environments. Our experiments show that, in terms of wall clock time, our JAX-based training pipeline is around 14 times faster than existing approaches. We also introduce and benchmark SMAX, a JAX-based approximate reimplementation of the popular StarCraft Multi-Agent Challenge.
  arXiv  Detail & Related papers  (2023-11-16T18:58:43Z)
• CoLA: Exploiting Compositional Structure for Automatic and Efficient Numerical Linear Algebra [62.37017125812101]
  We propose a simple but general framework for large-scale linear algebra problems in machine learning, named CoLA. By combining a linear operator abstraction with compositional dispatch rules, CoLA automatically constructs memory and runtime efficient numerical algorithms. We showcase its efficacy across a broad range of applications, including partial differential equations, Gaussian processes, equivariant model construction, and unsupervised learning.
  arXiv  Detail & Related papers  (2023-09-06T14:59:38Z)
• Target-independent XLA optimization using Reinforcement Learning [6.442130495735239]
  This paper introduces deep Reinforcement Learning based search for optimal XLA HLO pass ordering. We also propose enhancements to the deep RL algorithms to further improve optimal search performance. Overall, in our experimentation we observe an average of $13.3%$ improvement in operation count reduction.
  arXiv  Detail & Related papers  (2023-08-28T07:23:03Z)
• Harnessing Deep Learning and HPC Kernels via High-Level Loop and Tensor Abstractions on CPU Architectures [67.47328776279204]
  This work introduces a framework to develop efficient, portable Deep Learning and High Performance Computing kernels. We decompose the kernel development in two steps: 1) Expressing the computational core using Processing Primitives (TPPs) and 2) Expressing the logical loops around TPPs in a high-level, declarative fashion. We demonstrate the efficacy of our approach using standalone kernels and end-to-end workloads that outperform state-of-the-art implementations on diverse CPU platforms.
  arXiv  Detail & Related papers  (2023-04-25T05:04:44Z)
• Benchmarking the Linear Algebra Awareness of TensorFlow and PyTorch [1.1470070927586016]
  We develop benchmarks to investigate the linear algebra optimization capabilities of TF and PyT. In this work, we focus on linear algebra computations in TF and PyT.
  arXiv  Detail & Related papers  (2022-02-20T18:51:00Z)
• MLGO: a Machine Learning Guided Compiler Optimizations Framework [0.0]
  This work is the first full integration of machine learning in a complex compiler pass in a real-world setting. We use two different ML algorithms to train the inlining-for-size model, and achieve up to 7% size reduction. The same model generalizes well to a diversity of real-world targets, as well as to the same set of targets after months of active development.
  arXiv  Detail & Related papers  (2021-01-13T00:02:49Z)
• PolyScientist: Automatic Loop Transformations Combined with Microkernels for Optimization of Deep Learning Primitives [55.79741270235602]
  We develop a hybrid solution to the development of deep learning kernels. We use the advanced polyhedral technology to automatically tune the outer loops for performance.
  arXiv  Detail & Related papers  (2020-02-06T08:02:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.