Analyzing user behavior from server logs for improved virtualization management

Many organizations are shifting to virtualization of servers to improve operational efficiency and reduce capital expenditures. Virtualization helps consolidate servers, improve hardware utilization and reduce the consumption of power and physical space in the data center. However, the problem of tuning dynamic resource allocation is a novelty. Managing heterogeneous workloads running within virtual machines is an interesting and challenging topic of server virtualization. This paper presents an analysis of trending user access and their data size requirements, both of which are meaningful for virtualization management. The results suggest that user behaviors are different in terms of workload, day and time of usage. The data size requested by users also differs across time periods. This preliminary study may offer an approach to improve management of data centers running heterogeneous workloads using server virtualization.