Towards a Generic Application Partitioning and Retraction Framework for Pervasive Environments

TL;DR

This paper introduces BubbleCodes, a framework enabling context-aware applications to dynamically partition and retract across multiple computational nodes in pervasive environments, addressing the challenge of utilizing numerous small, powerful devices.

Contribution

The paper presents a novel, generic framework for application partitioning and retraction tailored for pervasive environments with multiple computational nodes.

Findings

Framework supports dynamic partitioning of applications.

Framework enables applications to retract and adapt in real-time.

Enhances utilization of small, powerful pervasive devices.

Abstract

Current mobile context-aware applications for pervasive environments have been designed to consume information from computational nodes or devices in their surroundings or environments. As the hardware industry continues making much smaller, compact and cheap hardware, the vision of having plenty of very small powerful digital networking nodes in, for e.g., the living room or bedroom, is not so far. Designing software that can make optimal use of all these computational nodes when needed is still challenging; since software will not only consume information from these nodes but parts of the software can be hosted on these different nodes. In this paper we propose the BubbleCodes Framework which is a generic application partitioning and retraction framework for next generation context-aware applications that will have the capabilities to partition and retract themselves on multiple computational nodes in a pervasive environment.