No abstract

When coupled with spatio-temporal context, location-based data collected in mobile cellular networks provide insights into patterns of human activity, interactions, and mobility. Whilst uncovered patterns have immense potential for improving services of telecom providers as well as for external applications related to social wellbeing, its inherent massive volume make such 'Big Data' sets complex to process. A significant number of studies involving such mobile phone data have been presented, but there still remain numerous open challenges to reach technology readiness. They include efficient access in privacy-preserving manner, high performance computing environments, scalable data analytics, innovative data fusion with other sources-all finally linked into the applications ready for operational mode. In this chapter, we provide a broad overview of the entire workflow from raw data access to the final applications and point out the critical challenges in each step that need to be addressed to unlock the value of data generated by mobile cellular networks.

Abstract. With the increasing architectural diversity of many-core architectures the challenges of parallel programming and code portability will sharply rise. The EU project PEPPHER addresses these issues with a component-based approach to application development on top of a taskparallel execution model. Central to this approach are multi-architectural components which encapsulate different implementation variants of application functionality tailored for different core types. An intelligent runtime system selects and dynamically schedules component implementation variants for efficient parallel execution on heterogeneous many-core architectures. On top of this model we have developed language, compiler and runtime support for a specific class of applications that can be expressed using the pipeline pattern. We propose C/C++ language annotations for specifying pipeline patterns and describe the associated compilation and runtime infrastructure. Experimental results indicate that with our high-level approach performance comparable to manual parallelization can be achieved.

No abstract

No abstract