Scheduling Algorithms for Efficient Execution of Stream Workflow Applications in Multicloud Environments

About

Big data processing applications are becoming more and more complex. They are no more monolithic in nature but instead they are composed of decoupled analytical processes in the form of a workflow. One type of such workflow applications is stream workflow application, which integrates multiple streaming big data applications to support decision making. Each analytical component of these applications runs continuously and processes data streams whose velocity will depend on several factors such as network bandwidth and processing rate of parent analytical component. As a consequence, the execution of these applications on cloud environments requires advanced scheduling techniques that adhere to end user's requirements in terms of data processing and deadline for decision making. In this paper, we propose two Multicloud scheduling and resource allocation techniques for efficient execution of stream workflow applications on Multicloud environments while adhering to workflow application and user performance requirements and reducing execution cost. Results showed that the proposed genetic algorithm is an adequate and effective for all experiments.

Mutaz Barika, Saurabh Garg, Andrew Chan, Rodrigo N. Calheiros · 2019

Related benchmarks

Task	Dataset	Result
Federated Learning Image Classification	CIFAR-10 IID	Accuracy99.38	15
Federated Learning Image Classification	CIFAR-10 (non-IID)	Convergence Accuracy98.5	15
Image Classification	Emnist-Digits IID D	Accuracy (Convergence)99.31	5
Image Classification	Emnist-Letters IID L	Convergence Accuracy92.8	5
Image Classification	CIFAR-10 (non-IID)	Convergence Accuracy54	5
Image Classification	Emnist-Digits Non-IID D	Convergence Accuracy98.6	5
Image Classification	Emnist-Letters Non-IID L	Accuracy80	5