Rapid SoC Prototyping Based on Virtual Multi-Processor Model

To meet both high performance and high energy efficiency, System-on-Chip (SoC) has a heterogenous architecture including many application-specific accelerators. As a design methodology for such heterogeneous SoCs, we propose a rapid SoC prototyping method based on virtual multi-processor model. Starting from a flexible, homogenous multi-processor model, the proposed method iteratively improves the energy efficiency by identifying inefficient portions of the design and replacing them with specialized hardware. Also, we present a SoC design environment Cyneum which we have been developing

[1]  David Black-Schaffer,et al.  Efficient Embedded Computing , 2008, Computer.

[2]  Krisztián Flautner,et al.  SoC-C: efficient programming abstractions for heterogeneous multicore systems on chip , 2008, CASES '08.

[3]  Scott A. Mahlke,et al.  Bridging the computation gap between programmable processors and hardwired accelerators , 2009, 2009 IEEE 15th International Symposium on High Performance Computer Architecture.

[4]  Vikram S. Adve,et al.  LLVM: a compilation framework for lifelong program analysis & transformation , 2004, International Symposium on Code Generation and Optimization, 2004. CGO 2004..

[5]  Daniel Gajski,et al.  A cycle-accurate compilation algorithm for custom pipelined datapaths , 2005, 2005 Third IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS'05).

[6]  Steven Swanson,et al.  Conservation cores: reducing the energy of mature computations , 2010, ASPLOS XV.

[7]  Christoforos E. Kozyrakis,et al.  Understanding sources of inefficiency in general-purpose chips , 2010, ISCA.