Optimal management of virtual infrastructures under flexible cloud service agreements

• Main Authors: GUO, Zhiling, LI, Jin, RAMESH, Ram
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2019
• Subjects: cloud computing; service level agreement (SLA); dynamic programming (DP); online algorithm; virtual machines (VMs); cloud resource management; Databases and Information Systems; Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/4828; https://ink.library.smu.edu.sg/context/sis_research/article/5831/viewcontent/ISR_Final_2019.pdf
• Institution: Singapore Management University

A cloud service agreement entails the provisioning of a required set of virtual infrastructure resources at a specified level of availability to a client. The agreement also lays out the price charged to the client and a penalty to the provider when the assured availability is not met. The availability assurance involves backup resource provisioning, and the provider needs to allocate backups cost-effectively by balancing the resource-provisioning costs with the potential penalty costs. We develop stochastic dynamic optimization models of the backup resource-provisioning problem, leading to cost-effective resource-management policies in different practical settings. We present two sets of dynamic provisioning strategies: periodic policies, where resources are adjusted at regular intervals, and aperiodic policies that allow flexible timing of such interventions. A closed-loop (CL) optimization model under conservative resource control and a certainty-equivalent (CE) optimization model under aggressive resource control are developed for periodic resource management. Similarly, aperiodic resource management is modeled by using two different strategies: single intervention with single look-ahead (SISL) and multiple interventions with single look-ahead (MISL). Online optimization algorithms for both the periodic and aperiodic models are developed. The worst-case behavior of the algorithms is studied by using competitive ratio analysis and the expected behavior by using computational investigations. By using these studies, managerial guidelines for choosing the best resource-management strategy under different client-specific, service-specific, and system-specific resource-optimization conditions are presented. We validate our models based on use cases constructed from Amazon Elastic Compute Cloud (EC2) with their actual pricing and service-credit data. The practical guidelines from this study will aid contract administrators in cloud data centers to both efficiently formulate service-level agreements and cost-effectively manage the virtual infrastructure resources committed in such agreements.