Trading-off system load and communication in mapping heuristics for improving NoC-based MPSoCs reliability

Improving embedded systems lifetime and reliability become a major concern for the semiconductor industry. Imbalanced mapping of applications may considerably impact on system lifetime since processors and NoC links located in hotspot zones may age faster than others, compromising the overall system performance. This work proposes a dynamic mapping heuristic that makes a trade-off between processors' load and NoC communication volume, aiming to increase system reliability. Results show the proposed heuristic provides a well-balanced workload distribution while reducing communication volume. Results showed that proposed mapping reduces application execution time (average 10%) and hotspots zones when compared to conventional mapping approaches.

[1]  Chao Chen,et al.  System-level reliability exploration framework for heterogeneous MPSoC , 2014, GLSVLSI '14.

[2]  Xiaobo Sharon Hu,et al.  Enhancing multicore reliability through wear compensation in online assignment and scheduling , 2013, 2013 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[3]  Bharadwaj Veeravalli,et al.  Temperature aware energy-reliability trade-offs for mapping of throughput-constrained applications on multimedia MPSoCs , 2014, 2014 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[4]  Vikas Chandra.  Quantifying workload dependent reliability in embedded processors , 2014, 2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC).

[5]  Fernando Gehm Moraes,et al.  Fast energy evaluation of embedded applications for many-core systems , 2014, 2014 24th International Workshop on Power and Timing Modeling, Optimization and Simulation (PATMOS).

[6]  Alexandre M. Amory,et al.  Multi-task dynamic mapping onto NoC-based MPSoCs , 2011, SBCCI '11.

[7]  Tei-Wei Kuo,et al.  Approximation Algorithms for Multiprocessor Energy-Efficient Scheduling of Periodic Real-Time Tasks with Uncertain Task Execution Time , 2008, 2008 IEEE Real-Time and Embedded Technology and Applications Symposium.

[8]  Bin Xie,et al.  An Efficient Power-Aware Optimization for Task Scheduling on NoC-based Many-core System , 2010, 2010 10th IEEE International Conference on Computer and Information Technology.

[9]  Donald E. Thomas,et al.  Lifetime improvement through runtime wear-based task mapping , 2012, CODES+ISSS '12.

[10]  Qiang Xu,et al.  Lifetime reliability-aware task allocation and scheduling for MPSoC platforms , 2009, 2009 Design, Automation & Test in Europe Conference & Exhibition.

[11]  Bharadwaj Veeravalli,et al.  Reliability-driven task mapping for lifetime extension of networks-on-chip based multiprocessor systems , 2013, 2013 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[12]  Bharadwaj Veeravalli,et al.  Run-time mapping for reliable many-cores based on energy/performance trade-offs , 2013, 2013 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFTS).

[13]  Donald E. Thomas,et al.  Cost-effective lifetime and yield optimization for NoC-based MPSoCs , 2014, TODE.

[14]  Fernando Gehm Moraes,et al.  Enhancing performance of MPSoCs through distributed resource management , 2012, 2012 19th IEEE International Conference on Electronics, Circuits, and Systems (ICECS 2012).