Publications:Meeting Engineer Efficiency Requirements in Highly Parallel Signal Processing by Using Platforms

From CERES
Jump to: navigation, search

Do not edit this section

Keep all hand-made modifications below

Title Meeting Engineer Efficiency Requirements in Highly Parallel Signal Processing by Using Platforms
Author Anders Åhlander and Anders Åström and Bertil Svensson and Mikael Taveniku
Year 2005
PublicationType Conference Paper
Journal
HostPublication Proceedings of the 17th IASTED International Conference on Parallel and Distributed Computing and Systems : November 14-16, 2005, Phoenix, AZ, USA
DOI
Conference International Conference on Parallel and Distributed Computing Systems, PDCS 2005, November 14-16, 2005, Phoenix, AZ, USA
Diva url http://hh.diva-portal.org/smash/record.jsf?searchId=1&pid=diva2:237590
Abstract One of the driving forces behind the development of new, highly parallel architectures is the need for embedded high-performance computing. The development of advanced applications on such architectures is, however, potentially connected to high costs. Cost-effective devel opment requires tools and processes that provide engineer efficiency, in this case tools and processes that help the developer master the application complexity. Related to engineer efficiency are the important concepts of system sustainability and flexibility. To address these issues, a platform approach can be taken. The platform should offer an understandable and stable development model, and at the same time give the possibility to take advantage of the rapid technology development, including the use of new parallel architectures. Thus it must support multiple hard ware targets, and the development model should decouple application development from mapping aspects. Two radar signal processing examples, one compute-intensive STAP and one data-intensive SAR, are used to illustrate the need. The GEPARD platform is presented as an example of our approach, and we argue that the described platform is a good fit for advanced signal processing development, facilitating the desired engineer efficiency, sustainability, and flexibility.