GIFT: A Coupon Based Throttle-and-Reward Mechanism for Fair and Efficient I/O Bandwidth Management on Parallel Storage Systems
Tirthak Patel, Rohan Garg, Devesh Tiwari
2020年份
27被引次数
2顶会引用
摘要
Large-scale parallel applications are highly data-intensive and perform terabytes of I/O routinely. Unfortunately, on a large-scale system where multiple applications run concurrently, I/O contention negatively affects system efficiency and causes unfair bandwidth allocation among applications. To address these challenges, this paper introduces GIFT, a principled dynamic approach to achieve fairness among competing applications and improve system efficiency.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper2
- Don't Forget the I/O When Allocating Your LLCYifan Yuan, Mohammad Alian, Yipeng Wang, Ren Wang 等ISCA 2021 · 被引用 37 次
- Fine-grained Policy-driven I/O Sharing for Burst BuffersEd Karrels, Lei Huang, Yuhong Kan, Ishank Arora 等SC 2023 · 被引用 5 次
相关 Paper
- Balancing efficiency and fairness in heterogeneous GPU clusters for deep learningShubham Chaudhary, Ramachandran Ramjee, Muthian Sivathanu, Nipun Kwatra 等EuroSys 2020 · 被引用 135 次
- Towards HPC I/O Performance Prediction through Large-scale Log AnalysisSunggon Kim, Alex Sim, Kesheng Wu, Suren Byna 等HPDC 2020 · 被引用 34 次
- R2B: high-efficiency and fair I/O scheduling for multi-tenant with differentiated demandsDiansen Sun, Yunpeng Chai, Chaoyang Liu, Weihao Sun 等DAC 2022 · 被引用 2 次
- Saba: Rethinking Datacenter Network Allocation from Application's PerspectiveM. R. Siavash Katebzadeh, Paolo Costa, Boris GrotEuroSys 2023 · 被引用 4 次
- Hitchhike: Efficient Request Submission via Deferred Enforcement of Address ContiguityXuda Zheng, Jian Zhou, Shuhan Bai, Runjin Wu 等ASPLOS 2026
