Surviving switch failures in cloud datacenters

Author:

Singh Rachee1,Mukhtar Muqeet1,Krishna Ashay1,Parkhi Aniruddha1,Padhye Jitendra1,Maltz David1

Affiliation:

1. Microsoft

Abstract

Switch failures can hamper access to client services, cause link congestion and blackhole network traffic. In this study, we examine the nature of switch failures in the datacenters of a large commercial cloud provider through the lens of survival theory. We study a cohort of over 180,000 switches with a variety of hardware and software configurations and find that datacenter switches have a 98% likelihood of functioning uninterrupted for over 3 months since deployment in production. However, there is significant heterogeneity in switch survival rates with respect to their hardware and software: the switches of one vendor are twice as likely to fail compared to the others. We attribute the majority of switch failures to hardware impairments and unplanned power losses. We find that the in-house switch operating system, SONiC, boosts the survival likelihood of switches in datacenters by 1% by eliminating switch failures caused by software bugs in vendor switch OSes.

Publisher

Association for Computing Machinery (ACM)

Subject

Computer Networks and Communications,Software

Reference12 articles.

1. Arista Networks. AAA Configuration. https://www.arista.com/en/um-eos/eos-aaa-configuration. (Accessed on 2020-05-11). Arista Networks. AAA Configuration. https://www.arista.com/en/um-eos/eos-aaa-configuration. (Accessed on 2020-05-11).

2. Arista Networks. EOS Central: Does this indicate a possible DRAM issue? https://eos.arista.com/forum/getting-ipt_crcerrpkt-and-jer_int_idr_mmu_ecc_1b_err_int-log-output-does-this-indicate-a-possible-dram-issue-on-bank-b/. (Accessed on 2020-05-11). Arista Networks. EOS Central: Does this indicate a possible DRAM issue? https://eos.arista.com/forum/getting-ipt_crcerrpkt-and-jer_int_idr_mmu_ecc_1b_err_int-log-output-does-this-indicate-a-possible-dram-issue-on-bank-b/. (Accessed on 2020-05-11).

3. Surviving failures in bandwidth-constrained datacenters

4. Regression Models and Life-Tables

5. Understanding network failures in data centers

Cited by 15 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Canary: Congestion-aware in-network allreduce using dynamic trees;Future Generation Computer Systems;2024-03

2. Photonic switched networking for data centers and advanced computing systems;Optical Fiber Communication Conference (OFC) 2024;2024

3. P4toNFV: Offloading from P4 switches to NFV in programmable data planes;International Journal of Communication Systems;2023-12-21

4. Physical Deployability Matters;Proceedings of the 22nd ACM Workshop on Hot Topics in Networks;2023-11-28

5. Availability modeling and evaluation of switches and data centers;2023 10th International Conference on Dependable Systems and Their Applications (DSA);2023-08-10

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3