Improving virtual machines networking performance for cloud computing

Virtualization is a key technology to enable cloud computing. It enhances resource availability and offers high flexibility and cost-effectiveness. However, the driver domain based model for network I/O virtualization exhibits poor networking performance. To overcome this limitation, we proposed a packet aggregation based I/O model in a previous work. In this paper, we studied the impact of the aggregation on the packets delay. For this purpose, we proposed a dimensioning tool based on the queueing theory modeling of the aggregation based I/O virtualization. The proposed tool allowed us to dynamically tune the aggregation mechanism in order to achieve the best tradeoff between the packets delay and throughput. This allows the cloud infrastructure providers to meet the user applications requirements.