2020/01/08 by Fabio López‐Pires, Lopez-Pires, Fabio, Lino Chamorro +3
Computer Science · #Cloud Computing and Resource Management #Distributed #FOS: Computer and information sciences #IoT and Edge/Fog Computing #Parallel #Software System Performance and Reliability #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.2001.02561
openalex publication_date 2020/01/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Cloud Service Brokers (CSBs) facilitate complex resource allocation\ndecisions, efficiently mapping dynamic tenant demands onto dynamic provider\noffers, where several objectives should ideally be considered. This work\nproposes for the first time a pure multi-objective formulation of a\nbroker-oriented Virtual Machine Placement (VMP) problem for dynamic\nenvironments, simultaneously optimizing the following objective functions: (i)\nTotal Infrastructure CPU (TICPU), (ii) Total Infrastructure Memory (TIMEM) and\n(iii) Total Infrastructure Price (TIP) while considering load balancing across\nproviders. To solve the formulated multi-objective problem, a Multi-Objective\nEvolutionary Algorithm (MOEA) is proposed. Considering that each time a demand\n(or offer) change occurs, a set of non-dominated solutions is found by\nPareto-based algorithms as the one proposed, different selection strategies\nwere evaluated in order to automatically select a convenient solution.\nAdditionally, the proposed algorithm, including the considered selection\nstrategies, was compared against mono-objective state-of-the-art alternatives\nin different scenarios with real data from providers in actual markets.\nExperimental results demonstrate that a pure multi-objective optimization\napproach considering the preferred solution selection strategy (S3)\noutperformed other mono-objective evaluated alternatives.\n