We describe and evaluate an Inter Cloud Manager (ICM) that dispatches jobs in Cloud environment. ICM consists of two main parts, Cloud monitoring and decision making. For Cloud monitoring, ICM use Cloud Probes (CPs) that observe and collect data, and decision making is based on both the measured execution time and network delay in forwarding the jobs and receiving back the result of the execution. ICM checks each Cloud hosts current number of waiting jobs and average execution time, and also makes use of delay and loss information regarding the network. Measurements are used to compare ICM with a Round Robin (RR) allocation of jobs between Clouds which spreads the workload equitably, and with a and Honeybee Foraging Algorithm (HFA). We see that under heavy load, ICM is better at avoiding system saturation than HFA and RR.
展开▼
