XLA: Difference between revisions
(Fix: remove --- horizontal lines (7 removed)) |
({{Status}}/{{TOC}} 템플릿 제거) |
||
| Line 1: | Line 1: | ||
== Overview == | == Overview == | ||
| Line 17: | Line 8: | ||
* 왜 필요한가? - HPC 및 서버 환경에서 필수 개념 | * 왜 필요한가? - HPC 및 서버 환경에서 필수 개념 | ||
* 언제 사용하는가? - 서버 구성, 성능 튜닝, 문제 해결 시 | * 언제 사용하는가? - 서버 구성, 성능 튜닝, 문제 해결 시 | ||
== Purpose == | == Purpose == | ||
| Line 26: | Line 16: | ||
* Scope: XLA의 개념, 사용법, 설정 | * Scope: XLA의 개념, 사용법, 설정 | ||
* Non-goals: 다른 주제로의 확장 | * Non-goals: 다른 주제로의 확장 | ||
== Key Concepts == | == Key Concepts == | ||
| Line 39: | Line 28: | ||
| [[Linux]], [[Server]] | | [[Linux]], [[Server]] | ||
|} | |} | ||
== Detailed Explanation == | == Detailed Explanation == | ||
| Line 45: | Line 33: | ||
[[파일:Openxla.png|섬네일|openxla]] | [[파일:Openxla.png|섬네일|openxla]] | ||
XLA<ref>https://github.com/openxla/xla</ref> (Accelerated Linear Algebra) is an open-source machine learning (ML) compiler for GPUs, CPUs, and ML accelerators. | XLA<ref>https://github.com/openxla/xla</ref> (Accelerated Linear Algebra) is an open-source machine learning (ML) compiler for GPUs, CPUs, and ML accelerators. | ||
The XLA compiler takes models from popular [[Deep Learning Frameworks|ML frameworks]] such as PyTorch, TensorFlow, and JAX, and optimizes them for high-performance execution across different hardware platforms including GPUs, CPUs, and ML accelerators. | The XLA compiler takes models from popular [[Deep Learning Frameworks|ML frameworks]] such as [[PyTorch]], TensorFlow, and JAX, and optimizes them for high-performance execution across different hardware platforms including GPUs, CPUs, and ML accelerators. | ||
In many cases, XLA improves performance over native TensorFlow. The major difference between these two is the fusion optimizer in XLA. Instead of executing many small kernels back to back, XLA optimizes these into larger kernels. This greatly reduces execution time of bandwidth bound kernels. XLA also offers many algebraic simplifications, far superior to what Tensorflow offers. | In many cases, XLA improves performance over native TensorFlow. The major difference between these two is the fusion optimizer in XLA. Instead of executing many small kernels back to back, XLA optimizes these into larger kernels. This greatly reduces execution time of bandwidth bound kernels. XLA also offers many algebraic simplifications, far superior to what Tensorflow offers. | ||
There are comprehensive instruction [https://docs.nvidia.com/deeplearning/frameworks/tensorflow-user-guide/index.html how to utilize XLA for TensorFlow] | There are comprehensive instruction [https://docs.nvidia.com/deeplearning/frameworks/tensorflow-user-guide/index.html how to utilize XLA for TensorFlow] | ||
<references/> | <references/> | ||
== Best Practices == | == Best Practices == | ||
| Line 56: | Line 43: | ||
* 공식 문서 참고 | * 공식 문서 참고 | ||
* 테스트 환경에서 먼저 검증 | * 테스트 환경에서 먼저 검증 | ||
== References == | == References == | ||
* [https://wiki.hpcmate.com XLA] | * [https://wiki.hpcmate.com XLA] | ||
== Related Pages == | == Related Pages == | ||
| Line 69: | Line 54: | ||
* [[Hardware]] | * [[Hardware]] | ||
* [[Network]] | * [[Network]] | ||
[[Category:Server]] | [[Category:Server]] | ||
Latest revision as of 13:28, 2 October 2026
Overview
XLA에 대한 기술 문서입니다.
Summary
- 무엇인가? - XLA
- 왜 필요한가? - HPC 및 서버 환경에서 필수 개념
- 언제 사용하는가? - 서버 구성, 성능 튜닝, 문제 해결 시
Purpose
이 문서가 존재하는 이유
- Goal: XLA에 대한 기술 정보 제공
- Scope: XLA의 개념, 사용법, 설정
- Non-goals: 다른 주제로의 확장
Key Concepts
| Concept | Description | Related |
|---|---|---|
| XLA | HPC/서버 환경에서 중요한 기술 개념 | Linux, Server |
Detailed Explanation
섬네일|openxla XLA[1] (Accelerated Linear Algebra) is an open-source machine learning (ML) compiler for GPUs, CPUs, and ML accelerators. The XLA compiler takes models from popular ML frameworks such as PyTorch, TensorFlow, and JAX, and optimizes them for high-performance execution across different hardware platforms including GPUs, CPUs, and ML accelerators. In many cases, XLA improves performance over native TensorFlow. The major difference between these two is the fusion optimizer in XLA. Instead of executing many small kernels back to back, XLA optimizes these into larger kernels. This greatly reduces execution time of bandwidth bound kernels. XLA also offers many algebraic simplifications, far superior to what Tensorflow offers. There are comprehensive instruction how to utilize XLA for TensorFlow
Best Practices
- 최신 버전 사용 권장
- 공식 문서 참고
- 테스트 환경에서 먼저 검증
References
Related Pages
Knowledge Graph
Related
→ AI Agents → NVIDIA GPU → CUDA → Transformer model → Transformer Engine