Corpus ID: 216868269

AIBench: An Industry Standard AI Benchmark Suite from Internet Services

@article{Tang2020AIBenchAI,
  title={AIBench: An Industry Standard AI Benchmark Suite from Internet Services},
  author={Fei Tang and Wanling Gao and Jianfeng Zhan and Chuanxin Lan and Xu Wen and Lei Wang and Chunjie Luo and Jiahui Dai and Zheng Cao and Xingwang Xiong and Zihan Jiang and Tianshu Hao and Fanda Fan and Fan Zhang and Yunyou Huang and Jianan Chen and Mengjia Du and Rui Ren and Chen Zheng and Daoyi Zheng and Haoning Tang and Kunlin Zhan and Biao Wang and Defei Kong and Minghe Yu and Chongkang Tan and Hu'an Li and Xinhui Tian and Yatao Li and Gang Lu and Junchao Shao and Zhenyu Wang and Xiao-yu Wang and Hainan Ye},
  journal={ArXiv},
  year={2020},
  volume={abs/2004.14690}
}
  • Fei Tang, Wanling Gao, +31 authors Hainan Ye
  • Published 2020
  • Computer Science
  • ArXiv
  • The booming successes of machine learning in different domains boost industry-scale deployments of innovative AI algorithms, systems, and architectures, and thus the importance of benchmarking grows. However, the confidential nature of the workloads, the paramount importance of the representativeness and diversity of benchmarks, and the prohibitive cost of training a state-of-the-art model mutually aggravate the AI benchmarking challenges. In this paper, we present a balanced AI benchmarking… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 63 REFERENCES

    Wait of a Decade: Did SPEC CPU 2017 Broaden the Performance Horizon?

    VIEW 2 EXCERPTS

    Fathom: reference workloads for modern deep learning methods

    VIEW 1 EXCERPT

    MLPerf Training Benchmark

    VIEW 3 EXCERPTS

    Memory Hierarchy for Web Search

    VIEW 1 EXCERPT

    In-datacenter performance analysis of a tensor processing unit

    VIEW 1 EXCERPT