arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2308.07541cs.DCcs.AIcs.SYeess.SY
收藏

无服务器计算中基于离线策略强化学习的按需冷启动频率降低

On-demand Cold Start Frequency Reduction with Off-Policy Reinforcement Learning in Serverless Computing

The University of Melbourne(墨尔本大学)

Siddharth Agarwal, Maria A. Rodriguez, Rajkumar Buyya

AI总结:

针对FaaS平台的冷启动延迟问题,提出基于无模型Q-learning的强化学习方法,根据CPU利用率和响应失败率等指标提前初始化函数,在Kubeless上实验表明吞吐量提升8.81%,计算负载和资源浪费分别降低55%和37%。

摘要:

Function-as-a-Service (FaaS) 是一种为应用程序提供事件驱动执行模型的云计算范式。它通过免除开发者的资源管理责任来体现无服务器特性,并为应用程序提供透明且按需的可扩展性。为了提供无缝的按需可扩展性,在函数实例不存在或不可用时,会准备新的函数实例来服务传入的工作负载。然而,FaaS 平台存在冷启动问题,这种函数准备过程会在函数响应中引入不可忽视的延迟,并降低最终用户体验。因此,本工作重点利用强化学习 (RL) 来减少平台上频繁的按需冷启动。所提出的方法使用无模型 Q-learning,综合考虑 CPU 利用率、现有函数实例和响应失败率等函数指标,根据预期需求提前主动初始化函数。该解决方案在 Kubeless 上实现,并使用应用于矩阵乘法函数的开源函数调用轨迹进行了评估。评估结果表明,与 Kubeless 的默认策略和函数保活策略相比,基于 RL 的智能体表现出更优的性能:吞吐量提高了 8.81%,计算负载和资源浪费分别减少了 55% 和 37%,这是冷启动减少的直接结果。

英文摘要:

Function-as-a-Service (FaaS) is a cloud computing paradigm offering an event-driven execution model to applications. It features serverless attributes by eliminating resource management responsibilities from developers, and offers transparent and on-demand scalability of applications. To provide seamless on-demand scalability, new function instances are prepared to serve the incoming workload in the absence or unavailability of function instances. However, FaaS platforms are known to suffer from cold starts, where this function provisioning process introduces a non-negligible delay in function response and reduces the end-user experience. Therefore, the presented work focuses on reducing the frequent, on-demand cold starts on the platform by using Reinforcement Learning(RL). The proposed approach uses model-free Q-learning that consider function metrics such as CPU utilization, existing function instances, and response failure rate, to proactively initialize functions, in advance, based on the expected demand. The proposed solution is implemented on Kubeless and evaluated using an open-source function invocation trace applied to a matrix multiplication function. The evaluation results demonstrate a favourable performance of the RL-based agent when compared to Kubeless' default policy and a function keep-alive policy by improving throughput by up to 8.81% and reducing computation load and resource wastage by up to 55% and 37%, respectively, that is a direct outcome of reduced cold starts.

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。

↑