AIBench: towards scalable and comprehensive datacenter AI benchmarking

W Gao, C Luo, L Wang, X Xiong, J Chen, T Hao… - … , and Optimizing: First …, 2019 - Springer
AI benchmarking provides yardsticks for benchmarking, measuring and evaluating
innovative AI algorithms, architecture, and systems. Coordinated by BenchCouncil, this …

Quantifying Resource Contention of Co-located Workloads with the System-level Entropy

Y Liang, S Zeng, L Wang - ACM Transactions on Architecture and Code …, 2023 - dl.acm.org
The workload co-location, such as deploying offline analysis workloads with online service
workloads on the same node, has become common for modern data centers. Workload co …

BOPS, not FLOPS! A new metric and roofline performance model for datacenter computing

L Wang, J Zhan, W Gao, KY Yang, ZH Jiang… - arXiv preprint arXiv …, 2018 - arxiv.org
For emerging datacenter (in short, DC) workloads, such as online Internet services or offline
data analytics, how to evaluate the upper bound performance and provide apple-to-apple …

Trace-based workload generation and execution

Y Sfakianakis, E Kanellou, M Marazakis… - Euro-Par 2021: Parallel …, 2021 - Springer
Although major cloud providers have captured and published workload executions in the
form of traces, it is not clear how to use them for workload generation on a wide range of …

Anomaly Detection Model for Process Resource Usage in Hybrid System based on eBPF and Isolation Forest

Z Zhang, L Chen - Proceedings of the 2023 6th International Conference …, 2023 - dl.acm.org
In the hybrid system, CPU-intensive processes, memory-intensive processes, and IO-
intensive processes can consume a significant amount of system resources, potentially …

BOPS, a new computation-centric metric for datacenter computing

L Wang, W Gao, K Yang, Z Jiang - … 2019, Denver, CO, USA, November 14 …, 2020 - Springer
For emerging datacenter (in short, DC) workloads, such as online Internet services or offline
data analytics, how to evaluate the upper bound performance and provide apple-to-apple …