Design Space Exploration Supporting Run-Time Resource Management - Multi-objective Design Space Exploration of Multiprocessor SoC Architectures - page 105

host processor will not cause too much overhead in terms of resources and execution
time (Chap. 9).

5.5 Conclusions

In this Chapter, we presented an automated tool flow which tackles Run-time
Resource Management challenge for multi-core embedded systems by efficiently
combining a design space exploration tool coupled with platform simulator(s). This
tool flow consisted of two steps: first a design-time heuristic methodology was
described which reduces number of configurations that an application can be run
optimally. This first design-time phase leverages optimization algorithms described
in Chap. 3 for minimizing the amount of simulations to be performed. Second step
is a light-weight run-time manager which selects an optimal configuration for each
running application depending on demanded QoS requirement. Hence, at run-time,
run-time manager leverages the design-time DSE results for deciding an operating
configuration to be loaded for each application. This operation is performed dynam-
ically, by following the QoS requirements of the specific use-case. Application of
this tool-flow on a real-life multimedia use case is described in Chap. 9 while other
frameworks operating at the Operating System (OS) level are presented in Chap. 6.

References

1. Avasare, P., Vanmeerbeeck, G., Kavka, C., Mariani, G.: Practical approach for design space
explorations using simulators at multiple abstraction levels. In: Design Automation Conference
(DAC) Users' Track (2010)
2. Baert, R., Brockmeyer, E., Wuytack, S., Ashby, T.: Exploring parallelizations of application
for mpsoc platforms using mpa. In: Proceedings of IEEE Design, Automation Test in Europe
Conference Exhibition (DATE), pp. 1148-1153. France (2009)
3. Baker, T.P.: An analysis of edf schedulability on a multiprocessor. IEEE Transactions on

Parallel and Distributed Systems 16 , 760-768 (2005)
4. Ballesi, P., Fornaciari, W., Siorpaes, D.: A hierarchical distributed control for power and per-
formances optimization of embedded systems. In: Conference on Architecture of Computing
Systems (ARCS), pp. 37-48. Hannover (2010)
5. Benini, L., Bogliolo, R., De Micheli, G.: A survey of design techniques for system-level
dynamic power management. IEEE Transactions on VLSI Systems 8 , 299-316 (2000)
6. Buchard, A.: Assigning real-time tasks to homogeneous multiprocessor systems. Technical

Report, University of Virginia (1994)
7. Cai, L., Gajski, D.: Transaction level modeling: an overview. In: Proceedings of CODES+ISSS
(2003)
8. Chan, H.L.: Non-migratory online deadline scheduling on multiprocessors. In: Proceedings
of SODA, pp. 970-979 (2004)
9. Chen, J.J., Yang, C.Y., Kuo, T.W., Shih, C.: Energy-efficient real-time task scheduling in
multiprocessor dvs systems. In: Design Automation Conference, 2007. ASP-DAC '07. Asia
and South Pacific, pp. 342-349 (2007)