PREACT: Predictive Resource Allocation for Bursty Workloads in a Co-located Data Center

• DOI: 10.1145/3673038.3673135
• Publication Date: 2024-08-08T22:29:01Z
• Authors: Dingyu Yang; Ziyang Xiao; Dongxiang Zhang; Shuhao Zhang; Jian Cao; Gang Chen
• Keywords: Data center; Service-level agreement; Service level objective; Resource Management

Co-locating online latency-critical (LC) services with best-effort (BE) batch jobs in the same server has been widely adopted by modern data centers to improve resource utilization. Various approaches have proposed maximize resources allocated BE without SLO (service level objective) violation. However, when facing bursty workloads, existing solutions suffer from poor performance because they cannot react promptly sudden and sharp increase of LC service requests. Consequently, these methods result either a high violation rate constraint or low utilization caused conservative allocation strategies.