UMP-PerComp: A Ubiquitous Multiprocessor Network-Based Pipeline Processing Framework for Pervasive Computing Environments

Pervasive computing provides an attractive vision for the future of computing where contextual intelligent services will be available anywhere at any time. However, how to integrate distributed resources existed among heterogeneous devices to build pervasive applications that can constantly adapt to the highly dynamic computing environment is still a challenge. By analyzing the pipelining feature within a user task and exploiting the parallelism among ubiquitous processors, we propose UMP-PerComp, a ubiquitous multiprocessor-based pipeline processing architecture to support high performance pervasive application development. UMP-PerComp is an integrated comprehensive framework that makes the development of pervasive applications easier. It includes a set of common services such as context management, resource management, service migration, topology management, that help to build applications and directly simplify the task decoupling and service composition. We describe the design and implementation of UMP-PerComp in this paper. Besides, we develop a distributed JPEG encoding application successfully on the UMP-PerComp. Our experience confirms that UMP-PerComp is a flexible, scalable and practicable framework with high performance.