arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.05155cs.DCcs.LG
收藏

面向数据中心软件升级的成本感知持续时间预测

Cost-aware Duration Prediction for Software Upgrades in Datacenters

Meta

Yi Ding, Aijia Gao, Thibaud Ryden, Michal Sedlak, Essam Ewaisha, Igor Marnat, Henry Hoffmann

AI总结:

本文提出Acela框架,通过考虑不对称预测成本和选择最佳模型,提升数据中心软件升级调度效率和吞吐量,实测提升升级窗口利用率1.25倍,升级数量增加33%。

摘要:

软件升级是维护数据中心服务器可靠性的重要环节。尽管作业持续时间预测和调度已广泛研究,但软件升级带来的独特挑战仍被低估。本文首次深入研究数据中心级别的软件升级调度。我们首先刻画各种升级类型,然后将调度任务建模为约束优化问题。为解决此问题,我们引入Acela,一种成本感知的持续时间预测框架,旨在提高升级调度效率和吞吐量,同时满足服务等级目标(SLOs)。Acela考虑不对称的预测成本,战略性地选择最佳预测模型,并缓解滞后效应导致的预测过高。在Meta生产数据中心系统的评估中,Acela显著提高了现有升级调度器的效率,通过提升升级窗口利用率1.25倍,增加计划和完成的升级数量33%和41%,并减少取消率2.4倍。代码和数据集将在论文通过后发布。

英文摘要:

Software upgrades are critical to maintaining server reliability in datacenters. While job duration prediction and scheduling have been extensively studied, the unique challenges posed by software upgrades remain largely under-explored. This paper presents the first in-depth investigation into software upgrade scheduling at datacenter scale. We begin by characterizing various types of upgrades and then frame the scheduling task as a constrained optimization problem. To address this problem, we introduce Acela, a cost-aware duration prediction framework designed to improve upgrade scheduling efficiency and throughput while meeting service-level objectives (SLOs). Acela accounts for asymmetric misprediction costs, strategically selects the best predictive models, and mitigates straggler-induced overestimations. Evaluations on Meta's production datacenter systems demonstrate that Acela significantly increases efficiency of the existing upgrade scheduler by improving upgrade window utilization by 1.25X, increasing the number of scheduled and completed upgrades by 33% and 41%, and reducing cancellation rates by 2.4X. The code and data sets will be released after paper acceptance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑