Optimal synthesis of latency and throughput constrained pipelined MPSoCs targeting streaming applications. Javaid, H., He, X., Ignjatovic, A., & Parameswaran, S. In Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2010 IEEE/ACM/IFIP International Conference on, pages 75-84, Oct, 2010.
abstract   bibtex   
A streaming application, characterized by a kernel that can be broken down into independent tasks which can be executed in a pipelined fashion, inherently allows its implementation on a pipeline of Application Specific Instruction set Processors (ASIPs), called a pipelined MPSoC. The latency and throughput requirements of streaming applications put constraints on the design of such a pipelined MPSoC, where each ASIP has a number of available configurations differing by additional instructions, and instruction and data cache sizes. Thus, the design space of a pipelined MPSoC is all the possible combinations of ASIP configurations (design points). In this paper, a methodology is proposed to optimize the area of a pipelined MPSoC under a latency or a throughput constraint. The final design point is a set of ASIP configurations with one configuration for each ASIP. We proposed an Integer Linear Programming (ILP) based solution to the area optimization problem under a latency constraint, and an algorithm for optimization of pipelined MPSoC area under a throughput constraint. The proposed solutions were evaluated using four streaming applications: JPEG encoder; JPEG decoder; MP3 encoder; and H.264 decoder. The time to find the Pareto front of each pipelined MPSoC was less than 4 minutes where design spaces had up to 1016 design points, illustrating the applicability of our approach.
@inproceedings{ 5751527,
  author = {Javaid, H. and Xin He and Ignjatovic, A. and Parameswaran, S.},
  booktitle = {Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2010 IEEE/ACM/IFIP International Conference on},
  title = {Optimal synthesis of latency and throughput constrained pipelined MPSoCs targeting streaming applications},
  year = {2010},
  month = {Oct},
  pages = {75-84},
  abstract = {A streaming application, characterized by a kernel that can be broken down into independent tasks which can be executed in a pipelined fashion, inherently allows its implementation on a pipeline of Application Specific Instruction set Processors (ASIPs), called a pipelined MPSoC. The latency and throughput requirements of streaming applications put constraints on the design of such a pipelined MPSoC, where each ASIP has a number of available configurations differing by additional instructions, and instruction and data cache sizes. Thus, the design space of a pipelined MPSoC is all the possible combinations of ASIP configurations (design points). In this paper, a methodology is proposed to optimize the area of a pipelined MPSoC under a latency or a throughput constraint. The final design point is a set of ASIP configurations with one configuration for each ASIP. We proposed an Integer Linear Programming (ILP) based solution to the area optimization problem under a latency constraint, and an algorithm for optimization of pipelined MPSoC area under a throughput constraint. The proposed solutions were evaluated using four streaming applications: JPEG encoder; JPEG decoder; MP3 encoder; and H.264 decoder. The time to find the Pareto front of each pipelined MPSoC was less than 4 minutes where design spaces had up to 1016 design points, illustrating the applicability of our approach.},
  keywords = {Pareto optimisation;application specific integrated circuits;integer programming;linear programming;multiprocessing systems;pipeline processing;system-on-chip;H.264 decoder;JPEG decoder;JPEG encoder;MP3 encoder;Pareto front;application specific instruction set processors;integer linear programming;optimal latency synthesis;streaming applications;throughput constrained pipelined MPSoC;Clocks;Optimization;Pipelines;Program processors;Streaming media;Throughput;Transform coding;Design Space Exploration;Integer Linear Programming}
}
Downloads: 0