2016/07/26 by Artur Baruchi, Edson Toshimi Midorikawa, Baruchi, Artur +5
Computer Science · #Cloud Computing and Resource Management #Distributed #FOS: Computer and information sciences #IoT and Edge/Fog Computing #Parallel #Peer-to-Peer Network Technologies #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.1607.07846
openalex publication_date 2016/07/26 · openalex created_date 2022/09/02 · openalex updated_date 2026/07/28
Virtual machine live migration in cloud environments aims at reducing energy\ncosts and increasing resource utilization. However, its potential has not been\nfully explored because of simultaneous migrations that may cause user\napplication performance degradation and network congestion. Research efforts on\nlive migration orchestration policies still mostly rely on system level\nmetrics. This work introduces an Application-aware Live Migration Architecture\n(ALMA) that selects suitable moments for migrations using application\ncharacterization data. This characterization consists in recognizing resource\nusage cycles via Fast Fourier Transform. From our experiments, live migration\ntimes were reduced by up to 74% for benchmarks and by up to 67% for real\napplications, when compared to migration policies with no application workload\nanalysis. Network data transfer during the live migration was reduced by up to\n62%.\n