Based On Cloud Computing Resource Constrained Environment Application Accelerated Research Framework

With the development of computing and communication, smart equipment with communication and computing capability has become quite widespread. But due to the limitations of the size, price and energy, the hardware resources (CPU, memory, etc.) of this smart equipment are quite limited. Therefore, smart equipment cannot execute the complex applications very well. Cloud computing is a research hotspot now. Taking the advantage of the mature virtualization technology, Cloud computing can manage and control the computing resources connected with the network as a whole, and provide them to the users as a service. Resource-constrained smart equipment can utilize the cloud computing resource to accelerate the execution of the local applications, which will improve the user experience and extended equipment application scenarios.Most existing cloud computing platforms are designed for the PC, they have complex programming framework and cannot achieve the dynamic migration of applications in the resource-constrained environment. Aiming at the resource-constrained application, this paper presents a cross-platform framework which can accelerate the execution of the application and have a simple programming model. The framework can move those complex tasks under the heterogeneous environment to the cloud, automatically and seamlessly, but leave those simple tasks executing in the local. As a result, we can speed up the execution of the entire application by accelerating complex tasks with the help of the abundant computing resources in the cloud. This framework adopts the serialization, virtualization and sandboxing, as well as distributed computing technology, and provides fine-grained computing migration strategy, simple user interface, good scalability and fault tolerance.Finally, we use orthogonal matching pursuit algorithm (Orthogonal Matching Pursuit OMP) in compressed sensing to take an application acceleration test in this framework. Experimental results suggest that the framework can accelerate the application in the resource-constrained environment.