The trend in computing systems is to combine various kinds of processing elements (PEs) to build more parallel architectures. This trend leads to more heterogeneous computing systems, for which abstractions are needed to efficiently program the systems without increasing the programming cost. This has lead to new programming languages and application programming interfaces (APIs). Parallel programming has always been a holy grail in computer science and dataflow programming promises a way to automatically provide parallel constructs for the programmer. This paper provides an approach to translate dataflow process networks (DPNs) into programs running some of the computations on the Open Computing Language (OpenCL) platform, supporting running programs on massively parallel hardware such as graphics processing units (GPUs). We show that certain DPN programs could run very efficiently on dataparallel architectures but also that there are certain patterns in DPN programs that prove problematic.

No abstract

The near channel performance of Low Density Parity Check Codes (LDPC) has motivated its wide applications. Iterative decoding of LDPC codes provides significant implementation challenges as the complexity grows with the code size. Recent trends in integrating Multiprocessor System on Chip (MPSoC) with Network on Chip (NoC) gives a modular platform for parallel implementation. This paper presents an implementation platform for decoding LDPC codes based on HeMPS, an open source MPSoC framework based on NoC communication fabric. Reduced minimum sum algorithm is used for decoding LDPC codes and simulations are performed using HeMPS tool. The data rate and speedup factor measured for decoding a rate 1/2 LDPC code characterised by 252 × 504 parity matrix is presented.

No abstract

No abstract