Active data structures on GPGPUs

O'Donnell, J. , Hall, C. and Monro, S. (2013) Active data structures on GPGPUs. In: The 6th Workshop on UnConventional High Performance Computing 2013 (UCHPC 2013), Aachen, Germany, 27 -30 Aug 2013,

Abstract

Active data structures support operations that may affect a large number of elements of an aggregate data structure. They are well suited for extremely fine grain parallel systems, including circuit parallelism. General purpose GPUs were designed to support regular graphics algorithms, but their intermediate level of granularity makes them potentially viable also for active data structures. We consider the characteristics of active data structures and discuss the feasibility of implementing them on GPGPUs. We describe the GPU implementations of two such data structures (ESF arrays and index intervals), assess their performance, and discuss the potential of active data structures as an unconventional programming model that can exploit the capabilities of emerging fine grain architectures such as GPUs.

• Item Type: Conference Proceedings
• Keywords: active data structure; data parallel; circuit parallel; GPU; GPGPU; index interval; functional array; selection
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Monro, Mr Stuart and O'Donnell, Dr John and Hall, Dr Cordelia
• Authors: O'Donnell, J., Hall, C., and Monro, S.
• Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science; Q Science > QA Mathematics > QA76 Computer software
• College/School: College of Science and Engineering > School of Computing Science
• Research Group: ENDS