Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2402.06194 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910477255180288 |
|---|---|
| author | Xiong, Yifan Jiang, Yuting Yang, Ziyue Qu, Lei Zhao, Guoshuai Liu, Shuguang Zhong, Dong Pinzur, Boris Zhang, Jie Wang, Yang Jose, Jithin Pourreza, Hossein Baxter, Jeff Datta, Kushal Ram, Prabhat Melton, Luke Chau, Joe Cheng, Peng Xiong, Yongqiang Zhou, Lidong |
| author_facet | Xiong, Yifan Jiang, Yuting Yang, Ziyue Qu, Lei Zhao, Guoshuai Liu, Shuguang Zhong, Dong Pinzur, Boris Zhang, Jie Wang, Yang Jose, Jithin Pourreza, Hossein Baxter, Jeff Datta, Kushal Ram, Prabhat Melton, Luke Chau, Joe Cheng, Peng Xiong, Yongqiang Zhou, Lidong |
| contents | Reliability in cloud AI infrastructure is crucial for cloud service providers, prompting the widespread use of hardware redundancies. However, these redundancies can inadvertently lead to hidden degradation, so called "gray failure", for AI workloads, significantly affecting end-to-end performance and concealing performance issues, which complicates root cause analysis for failures and regressions.
We introduce SuperBench, a proactive validation system for AI infrastructure that mitigates hidden degradation caused by hardware redundancies and enhances overall reliability. SuperBench features a comprehensive benchmark suite, capable of evaluating individual hardware components and representing most real AI workloads. It comprises a Validator which learns benchmark criteria to clearly pinpoint defective components. Additionally, SuperBench incorporates a Selector to balance validation time and issue-related penalties, enabling optimal timing for validation execution with a tailored subset of benchmarks. Through testbed evaluation and simulation, we demonstrate that SuperBench can increase the mean time between incidents by up to 22.61x. SuperBench has been successfully deployed in Azure production, validating hundreds of thousands of GPUs over the last two years. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_06194 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SuperBench: Improving Cloud AI Infrastructure Reliability with Proactive Validation Xiong, Yifan Jiang, Yuting Yang, Ziyue Qu, Lei Zhao, Guoshuai Liu, Shuguang Zhong, Dong Pinzur, Boris Zhang, Jie Wang, Yang Jose, Jithin Pourreza, Hossein Baxter, Jeff Datta, Kushal Ram, Prabhat Melton, Luke Chau, Joe Cheng, Peng Xiong, Yongqiang Zhou, Lidong Distributed, Parallel, and Cluster Computing Reliability in cloud AI infrastructure is crucial for cloud service providers, prompting the widespread use of hardware redundancies. However, these redundancies can inadvertently lead to hidden degradation, so called "gray failure", for AI workloads, significantly affecting end-to-end performance and concealing performance issues, which complicates root cause analysis for failures and regressions. We introduce SuperBench, a proactive validation system for AI infrastructure that mitigates hidden degradation caused by hardware redundancies and enhances overall reliability. SuperBench features a comprehensive benchmark suite, capable of evaluating individual hardware components and representing most real AI workloads. It comprises a Validator which learns benchmark criteria to clearly pinpoint defective components. Additionally, SuperBench incorporates a Selector to balance validation time and issue-related penalties, enabling optimal timing for validation execution with a tailored subset of benchmarks. Through testbed evaluation and simulation, we demonstrate that SuperBench can increase the mean time between incidents by up to 22.61x. SuperBench has been successfully deployed in Azure production, validating hundreds of thousands of GPUs over the last two years. |
| title | SuperBench: Improving Cloud AI Infrastructure Reliability with Proactive Validation |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2402.06194 |