SATORI: Efficient and Fair Resource Partitioning by Sacrificing Short-Term Benefits for Long-Term Gains*
Rohan Basu Roy, Tirthak Patel, Devesh Tiwari
Abstract
Multi-core architectures have enabled data centers to increasingly co-locate multiple jobs to improve resource utilization and lower the operational cost. Unfortunately, naively co-locating multiple jobs may lead to only a modest increase in system throughput. Worse, some users may observe proportionally higher performance degradation compared to other users co-located on the same physical multi-core system. SATORI is a novel strategy to partition multi-core architectural resources to achieve two conflicting goals simultaneously: increasing system throughput and achieving fairness among the co-located jobs.
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Cited by top-tier papers6
- RIBBON: cost-effective and qos-aware deep learning model inference using a diverse pool of cloud computing instancesBaolin Li, Rohan Basu Roy, Tirthak Patel, Vijay Gadepally et al.SC 2021 · 16 citations
- SCOOT: SLO-Oriented Performance Tuning for LLM Inference EnginesKe Cheng, Zhi Wang, Wen Hu, Tiannuo Yang et al.WWW 2025 · 13 citations
- ESG: Pipeline-Conscious Efficient Scheduling of DNN Workflows on Serverless Platforms with Shareable GPUsXinning Hui, Yuanchao Xu, Zhishan Guo, Xipeng ShenHPDC 2024 · 11 citations
- Mosaic: Harnessing the Micro-Architectural Resources of Servers in Serverless EnvironmentsJovan Stojkovic, Esha Choukse, Enrique Saurez, Íñigo Goiri et al.MICRO 2024 · 6 citations
- Litmus: Fair Pricing for Serverless ComputingQi Pei, Yipeng Wang, Seunghee ShinASPLOS 2024 · 1 citation
Related papers
- CLITE: Efficient and QoS-Aware Co-Location of Multiple Latency-Critical Jobs for Warehouse Scale ComputersTirthak Patel, Devesh TiwariHPCA 2020 · 153 citations
- OLPart: Online Learning based Resource Partitioning for Colocating Multiple Latency-Critical Jobs on Commodity ComputersRuobing Chen, Haosen Shi, Yusen Li, Xiaoguang Liu et al.EuroSys 2023 · 27 citations
- DRLPart: A Deep Reinforcement Learning Framework for Optimally Efficient and Robust Resource Partitioning on Commodity ServersRuobing Chen, Jinping Wu, Haosen Shi, Yusen Li et al.HPDC 2021 · 22 citations
- Holmes: SMT Interference Diagnosis and CPU Scheduling for Job Co-locationAidi Pi, Xiaobo Zhou, Chengzhong XuHPDC 2022 · 11 citations
- Rhythm: component-distinguishable workload deployment in datacentersLaiping Zhao, Yanan Yang, Kaixuan Zhang, Xiaobo Zhou et al.EuroSys 2020 · 49 citations
