PROJECT TITLE :

ANGEL: Agent-Based Scheduling for Real-Time Tasks in Virtualized Clouds

ABSTRACT:

The success of Cloud Computing makes an increasing variety of real-time applications like Signal Processing and Weather Forecasting run in the cloud. Meanwhile, scheduling for real-time tasks is taking part in an essential role for a cloud supplier to maintain its quality of service and enhance the system’s performance. During this paper, we devise a novel agent-primarily based scheduling mechanism in cloud computing atmosphere to allocate real-time tasks and dynamically provision resources. In distinction to ancient contract net protocols, we employ a bidirectional announcement-bidding mechanism and also the collaborative process consists of 3 phases, i.e., basic matching part, forward announcement-bidding phase and backward announcement-bidding section. Moreover, the elasticity is sufficiently thought-about whereas scheduling by dynamically adding virtual machines to improve schedulability. Furthermore, we style calculation rules of the bidding values in both forward and backward announcement-bidding phases and 2 heuristics for selecting contractors. On the premise of the bidirectional announcement-bidding mechanism, we have a tendency to propose an agent-primarily based dynamic scheduling algorithm named ANGEL for real-time, freelance and aperiodic tasks in clouds. Intensive experiments are conducted on CloudSim platform by injecting random artificial workloads and also the workloads from the last version of the Google cloud tracelogs to evaluate the performance of our ANGEL. The experimental results indicate that ANGEL will efficiently solve the real-time task scheduling drawback in virtualized clouds.