Libra: Harvesting Idle Resources Safely and Timely in Serverless Clusters
Libra: Harvesting Idle Resources Safely and Timely in Serverless Clusters
复制标题
DOI：
10.1145/3588195.3592996
复制
发表时间：
2023-08
期刊：
影响因子：
--
通讯作者：
Hanfei Yu;Christian Fontenot;Hao Wang;Jian Li;Xu Yuan;Seung-Jong Park
中科院分区：
文献类型：
--
作者：
Hanfei Yu;Christian Fontenot;Hao Wang;Jian Li;Xu Yuan;Seung-Jong Park
无服务器计算受到了来自各个行业的用户和基础设施提供商的青睐，包括在线服务和科学计算。用户享受其自动扩展和易于管理，提供商拥有更多的控制权来优化他们的服务。然而，现有的无服务器平台仍然需要用户为其功能预先定义资源分配，导致实践中经验不足的用户经常错误配置。此外，功能不断变化的输入数据进一步扩大了其动态资源需求和静态分配之间的差距，使功能要么过度配置，要么配置不足。本文介绍了Libra，这是一个用于多节点无服务器集群的安全及时的资源收集框架。Libra做出精确的收获决策，通过主动分析动态资源需求和可用性来加速收获资源的函数调用，并共同提高资源利用率。在OpenWhisk集群上进行的实验表明，与最先进的解决方案相比，Libra将响应延迟降低了39%，并实现了3倍的资源利用率。
Serverless computing has been favored by users and infrastructure providers from various industries, including online services and scientific computing. Users enjoy its auto-scaling and ease-of-management, and providers own more control to optimize their service. However, existing serverless platforms still require users to pre-define resource allocations for their functions, leading to frequent misconfiguration by inexperienced users in practice. Besides, functions' varying input data further escalate the gap between their dynamic resource demands and static allocations, leaving functions either over-provisioned or under-provisioned. This paper presents Libra, a safe and timely resource harvesting framework for multi-node serverless clusters. Libra makes precise harvesting decisions to accelerate function invocations with harvested resources and jointly improve resource utilization by profiling dynamic resource demands and availability proactively. Experiments on OpenWhisk clusters with real-world workloads show that Libra reduces response latency by 39% and achieves 3X resource utilization compared to state-of-the-art solutions.