Exploration methodology of dynamic data structures in multimedia and network applications for embedded platforms

Abstract

In the last years, there is a trend towards network and multimedia applications to be implemented in portable devices. These applications usually contain complex dynamic data structures. The appropriate selection of the dynamic data type (DDT) combination of an application affects the performance and the energy consumption of the whole system. Thus, DDT exploration methodology is used to perform trade-offs between design factors, such as performance and energy consumption. In this paper we provide a new approach to the DDT exploration procedure, based on a new library of DDTs which remedies the limitations of an existing solution and allows the DDT optimization of a wider range of application domains. Using the new library, we performed DDT exploration in network and multimedia benchmarks and achieved performance and energy consumption improvements up to 22% and 5.8%, respectively.