A Network-aware Virtual Machine Placement and Migration Approach in Cloud Computing

TLDR

Cloud computing delivers on‑demand shared resources, but data‑intensive applications rely heavily on network I/O, making VM placement and migration critical for performance. The authors aim to reduce data transfer time by proposing a virtual machine placement and migration strategy. Their strategy is network‑aware, selecting VM locations and migrating them to minimize data transfer time consumption. Simulation results demonstrate that the approach effectively optimizes data transfer between VMs and data, thereby improving overall application performance.

Abstract

Cloud computing represents a major step up in computing whereby shared computation resources are provided on demand. In such a scenario, applications and data thereof can be hosted by various networked virtual machines (VMs). As applications, especially data-intensive applications, often need to communicate with data frequently, the network I/O performance would affect the overall application performance significantly. Therefore, placement of virtual machines which host an application and migration of these virtual machines while the unexpected network latency or congestion occurs is critical to achieve and maintain the application performance. To address these issues, this paper proposes a virtual machine placement and migration approach to minimizing the data transfer time consumption. Our simulation studies suggest that the proposed approach is effective in optimizing the data transfer between the virtual machine and data, thus helping optimize the overall application performance.