Despite the progress made in developing more advanced compilers for embedded systems, programming of embedded high-performance computing systems based on Digital Signal Processors (DSPs) is still a highly skilled manual task. This is true for single-processor systems, and even more for embedded systems based on multiple DSPs. Compilers often fail to optimise existing DSP codes written in C due to the employed programming style. Parallelisation is hampered by the complex multiple address space memory architecture, which can be found in most commercial multi-DSP configurations. This thesis develops an integrated optimisation and parallelisation strategy that can deal with low-level C codes and produces optimised parallel code for a homogeneous multi-DSP architecture with distributed physical memory and multiple logical address spaces. In a first step, low-level programming idioms are identified and recovered. This enables the application of high-level code and data transformations well-known in the field of scientific computing. Iterative feedback-driven search for “good” transformation sequences is being investigated. A novel approach to parallelisation based on a unified data and loop transformation framework is presented and evaluated. Performance optimisation is achieved through exploitation of data locality on the one hand, and utilisation of DSP-specific architectural features such as Direct Memory Access (DMA) transfers on the other hand. The proposed methodology is evaluated against two benchmark suites (DSPstone & UTDSP) and four different high-performance DSPs, one of which is part of a commercial four processor multi-DSP board also used for evaluation. Experiments confirm the effectiveness of the program recovery techniques as enablers of high-level transformations and automatic parallelisation. Source-to-source transformations of DSP codes yield an average speedup of 2.21 across four different DSP architectures. The parallelisation scheme is – in conjunction with a set of locality optimisations – able to produce linear and even super-linear speedups on a number of relevant DSP kernels and applications.

<p>In communication systems, the need for high bandwidth interconnects and</p><p>efficient distribution of large amount of data is very essential. This thesis work</p><p>addresses all-optical packet switching issues in the field of reconfigurable optical</p><p>interconnection networks for high performance embedded systems. The recent</p><p>research conducted at the Halmstad University, on high performance embedded</p><p>systems, focuses on the optical interconnection techniques to achieve ultra high</p><p>throughputs and reconfigurability at the system level.</p><p>Recent research in the field of optical interconnection networks for applications</p><p>like switches and routers for data and telecommunication industry and parallel</p><p>computing architectures for embedded signal processing use optical to electrical</p><p>conversion to switch packets. This conversion scales down the enormous bandwidth</p><p>capacity of the optical communication channels to electronic processing rates. To</p><p>maintain the high throughputs all over the interconnection networks, the optical</p><p>packets need to be maintained in optical state and switched to different part of the</p><p>interconnection network. To achieve this goal, all-optical packet switching</p><p>architectures are studied. The study is concluded with a positive outlook towards alloptical</p><p>switching technologies, and it will play a very important role in the near</p><p>future in the field of optical communication, telecommunication and embedded</p><p>systems.</p>

In communication systems, the need for high bandwidth interconnects and efficient distribution of large amount of data is very essential. This thesis work addresses all-optical packet switching issues in the field of reconfigurable optical interconnection networks for high performance embedded systems. The recent research conducted at the Halmstad University, on high performance embedded systems, focuses on the optical interconnection techniques to achieve ultra high throughputs and reconfigurability at the system level. Recent research in the field of optical interconnection networks for applications like switches and routers for data and telecommunication industry and parallel computing architectures for embedded signal processing use optical to electrical conversion to switch packets. This conversion scales down the enormous bandwidth capacity of the optical communication channels to electronic processing rates. To maintain the high throughputs all over the interconnection networks, the optical packets need to be maintained in optical state and switched to different part of the interconnection network. To achieve this goal, all-optical packet switching architectures are studied. The study is concluded with a positive outlook towards alloptical switching technologies, and it will play a very important role in the near future in the field of optical communication, telecommunication and embedded systems.