Runtime QoS support for MPSoC: A processor centric approach

The MPSoC literature related to runtime support of Quality of Service (QoS) presents proposals related to the management of the interconnection infrastructure and the processing elements (PEs). The QoS management of computation resources is essential to fulfill real-time (RT) applications, as in multimedia systems, where dynamic workload and CPU sharing are commonplace. However, few works concerning QoS at the processor level for RT applications are found in the literature. The proposed work provides a runtime support for QoS acting in the PEs, coupled to a monitoring scheme at the task level. The main goal of the present work is to employ a low overhead task migration combined with task scheduling priority, to increase the computation resources for RT applications. An important feature of the proposal is to act indirectly over the QoS application, by minimizing the interference of the best effort (BE) tasks in the RT application performance. If the monitoring infrastructure still detects deadline misses, the system management then tries to optimize the RT application acting at the task level, migrating the affected RT task or modifying the scheduling policy. The NoC-based MPSoC was modeled and validated using an RTL description, with real applications. Results use throughput as the reference performance parameter. The proposed technique restored the RT applications performance after the introduction of disturbing applications, with a small reaction time.

[1]  Qing Li,et al.  Real-Time Concepts for Embedded Systems , 2003 .

[2]  Xiaola Lin,et al.  Deadlock-Free Multicast Wormhole Routing in 2-D Mesh Multicomputers , 1994, IEEE Trans. Parallel Distributed Syst..

[3]  Hokeun Kim,et al.  A task remapping technique for reliable multi-core embedded systems , 2010, 2010 IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS).

[4]  Federico Angiolini,et al.  QoS-Driven Reconfigurable Parallel Computing for NoC-Based Clustered MPSoCs , 2013, IEEE Transactions on Industrial Informatics.

[5]  Fernando Gehm Moraes,et al.  Energy-aware dynamic task mapping for NoC-based MPSoCs , 2011, 2011 IEEE International Symposium of Circuits and Systems (ISCAS).

[6]  Michel Robert,et al.  MPI-Based Adaptive Task Migration Support on the HS-Scale System , 2008, 2008 IEEE Computer Society Annual Symposium on VLSI.

[7]  Wei Quan,et al.  A scenario-based run-time task mapping algorithm for MPSoCs , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[8]  Zhe Ma,et al.  Run-time parallelization switching for resource optimization on an MPSoC platform , 2014, Des. Autom. Embed. Syst..

[9]  Fernando Gehm Moraes,et al.  Differentiated Communication Services for NoC-Based MPSoCs , 2014, IEEE Trans. Computers.

[10]  Fernando Gehm Moraes,et al.  Proposal and evaluation of a task migration protocol for NoC-based MPSoCs , 2012, 2012 IEEE International Symposium on Circuits and Systems.

[11]  Simon Holmbacka,et al.  Task Migration for Dynamic Power and Performance Characteristics on Many-Core Distributed Operating Systems , 2013, 2013 21st Euromicro International Conference on Parallel, Distributed, and Network-Based Processing.

[12]  Chifeng Wang,et al.  Design and Evaluation of a High Throughput QoS-Aware and Congestion-Aware Router Architecture for Network-on-Chip , 2012, 2012 20th Euromicro International Conference on Parallel, Distributed and Network-based Processing.

[13]  George Kornaros,et al.  A survey and taxonomy of on-chip monitoring of multicore systems-on-chip , 2013, TODE.

[14]  Sergio Johann Filho,et al.  Task model suitable for dynamic load balancing of real-time applications in NoC-based MPSoCs , 2012, 2012 IEEE 30th International Conference on Computer Design (ICCD).

[15]  George Kornaros,et al.  Hardware-assisted dynamic power and thermal management in multi-core SoCs , 2011, GLSVLSI '11.

[16]  Yuping Zhang,et al.  Workload-balancing schedule with adaptive architecture of MPSoCs for fault tolerance , 2010, 2010 3rd International Conference on Biomedical Engineering and Informatics.

[17]  Paolo Meloni,et al.  Adaptivity Support for MPSoCs Based on Process Migration in Polyhedral Process Networks , 2012, VLSI Design.

[18]  Hamid Noori,et al.  Proactive task migration with a self-adjusting migration threshold for dynamic thermal management of multi-core processors , 2014, The Journal of Supercomputing.

[19]  Rached Tourki,et al.  A Quality of Service Network on Chip based on a new priority arbitration mechanism , 2011, ICM 2011 Proceeding.