Abstract
Since the application complexity is growing and applications can be dynamically activated, the major challenge for heterogeneous multiprocessor platforms is to select at runtime an energy-efficient mapping of these applications. Taking into account that many different possible implementations per application can be available, and that the selection must meet the application deadlines under the available platform resources, this runtime optimization problem can be modeled as a Multidimension Multichoice Knapsack Problem (MMKP), which is known to be NP-hard. Not only algorithms for an optimal solution, but also state-of-the-art heuristics for real-time systems are still too slow for runtime management of multiprocessor platforms. This article provides a new fast and lightweight heuristic for finding near-optimal solutions for MMKP problems. The main contribution of this heuristic is: (i) the Pareto filtering of each initial MMKP set to reduce the search space, (ii) the sorting of all Pareto points together in a single two-dimension search space, where (iii) a very fast greedy algorithm solves the MMKP. Experiments show that our heuristic finds solutions close (within 0% to 0.4%) to the ones obtained by the fastest state-of-the-art heuristics, in just a fraction of the execution time (more than 97.5% gain on a StrongARM processor) and can run in less than 1ms for multiprocessor problem sizes. This is required for realistic OS reaction times in video and wireless application sets.
- Akbar, M., Manning, E., Shoja, G., and Khan, S. 2001. Heuristic solutions for the multiple-choice multiple-dimension knapsack problem. In Proceedings of the Conference on Computational Science. Springer, Berlin. Google Scholar
Digital Library
- Akbar, M., Rahman, M., Kaykobad, M., Manning, E., and Shoja, G. 2006. Solving the multi-dimensional multiple-choice knapsack problem by constructing convex hulls. Comput. Oper. Res. 33, 5, 1259--1273.Google Scholar
Cross Ref
- Armstrong, R., Kung, D., Sinha, P., and Zoltners, A. 1983. A computational study of multiple-choice knapsack algorithm. ACM Trans. Math. Softw. 9, 184--198. Google Scholar
Digital Library
- Bougard, B., Pollin, S., Lenoir, G., Van der Perre, L., Catthoor, F., and Dehaene, W. 2004. Transport level performance-energy trade-off in wireless networks and consequences on the system-level architecture and design paradigm. In Proceedings of the Workshop on Signal Processing Systems. IEEE, Los Alamitos, CA, 77--82.Google Scholar
- CERMSEM. 2004. Multi-choice multi-dimension knapsack problems. ftp://cermsem.univ-paris1.fr/pub/CERMSEM/hi_/MMKP/MMKP.html.Google Scholar
- Cormen, T., Leiserson, C., and Rivest, R. 1990. Introduction to Algorithms. MIT Press, Cambridge, MA. Google Scholar
Digital Library
- Dammeyer, F. and Voss, S. 1991. Dynamic tabu list management using the reverse elimination method. Ann. Oper. Res. Google Scholar
Digital Library
- Drexel, A. 1988. A simulated annealing approach to the multi-constraint zero-one knapsack problem. Ann. Comput. 40, 1--8. Google Scholar
Digital Library
- Khan, S. 1998. Quality adaptation in a multi-session adaptive multimedia system: Model and architecture. Ph.D. thesis, Department of Electrical and Computer Engineering, University of Victoria. Google Scholar
Digital Library
- Khan, S., Li, K., and Manning, E. 1997. The utility model for adaptive multimedia systems. In Proceedings of the International Workshop on Multimedia Modeling. 111--126.Google Scholar
- Khan, S., Li, K., Manning, E., and Akbar, M. 2002. Solving the knapsack problem for adaptive multimedia system. Studia Informatica Universalis, 161--182.Google Scholar
- Khuri, S., Back, T., and Heitkotter, J. 1994. The zero/one multiple knapsack problem and genetic algorithms. InProceedings of the Symposium of Applied Computation. ACM, New York. Google Scholar
Digital Library
- Koleser, P. 1967. A branch and bound algorithm for knapsack problem. Manage. Sci. 13, 723--735.Google Scholar
- Lee, C., Lehoczky, J., Siewiorek, D., Rajkumar, R., and Hansen, J. 1999. A scalable solution to the multi-resource QoS problem. In Proceedings of the Real-Time Systems Symposium. IEEE, Los Alamitos, CA, 315--326. Google Scholar
Digital Library
- Mangharam, R., Pollin, S., Bougard, B., Rajkumar, R., Catthoor, F., Van der Perre, L., and Moerman, I. 2005. Optimal fixed and scalable energy management for wireless networks. In Proceedings of the Conference on Computer Communications. IEEE, Los Alamitos, CA.Google Scholar
- Martello, S. and Toth, P. 1990. Knapsack Problems: Algorithms and Computer Implementations. J. Wiley and Sons, Hoboken, NJ. Google Scholar
Digital Library
- Maruyama, T., Takagi, M., and Hoshino, T. 1999. Hardware implementation techniques for recursive calls and loops. In Proceedings of the International Conference on Field Programmable Logic and Applications. IEEE, Los Alamitos, CA, 450--455. Google Scholar
Digital Library
- McVoy, L. W. and Staelin, C. 1996. lmbench: Portable tools for performance analysis. In Proceedings of the Annual Technical Conference. USENIX, Berkeley, CA, 279--294. Google Scholar
Digital Library
- Mejia-Alvarez, P., Levner, E., and Mosse, D. 2002. Power-optimized scheduling server for real-time tasks. In Proceedings of the Real-Time and Embedded Technology and Applications Symposium. IEEE, Los Alamitos, CA, 239--250. Google Scholar
Digital Library
- Moser, M., Jokanovic, D., and Shiratori, N. 1997. An algorithm for the multi-dimensional multiple-choice knapsack problem. IEICE Tran. Fundam. Electron. 80, 3, 582--589.Google Scholar
- Parra-Hernandez, R. and Dimopoulos, N. 2005. A new heuristic for solving the multi-choice multi-dimensional knapsack problem. IEEE Trans. Syst. Man Cybernetics Part A Syst. Humans 35, 5, 708--717. Google Scholar
Digital Library
- Qin, W. and Malik, S. 2003. Flexible and formal modeling of micro-processors with application to retargetable simulation. In Proceedings of the Conference on Design, Automation and Test in Europe. IEEE, Los Alamitos, CA, 556--561. Google Scholar
Digital Library
- Tachibana, T., Murata, Y., Shibata, N., Yasumoto, K., and Ito, M. 2006. Flexible implementation of genetic algorithms on fpgas. In Proceedings of the International Symposium on Field Programmable Gate Arrays. ACM, New York, 236. Google Scholar
Digital Library
- Volder, J. E. 1959. The CORDIC trigonometric computing technique. IRE Trans. Electron. Comput. EC-8, 330--334.Google Scholar
- Yang, P. and Catthoor, F. 2003. Pareto-optimization-based run-time task scheduling for embedded systems. In Proceedings of the International Symposium on System Synthesis. IEEE, Los Alamitos, CA, 120--125. Google Scholar
Digital Library
- Ykman-Couvreur, C., Brockmeyer, E., Nollet, V., Marescaux, T., Catthoor, F., and Corporaal, H. 2005. Design-time application exploration for MP-SoC customized run-time management. In Proceedings of the International Symposium on System-on-Chip. IEEE, Los Alamitos, CA, 66--73.Google Scholar
- Ykman-Couvreur, C., Nollet, V., Marescaux, T., Brockmeyer, E., Catthoor, F., and corporaal, H. 2006. Pareto-based application specification for MP-SoC customized run-time management. In Proceedings of the International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation. IEEE, Los Alamitos, CA, 78--84.Google Scholar
Cross Ref
Index Terms
Fast multidimension multichoice knapsack heuristic for MP-SoC runtime management
Recommendations
A New Heuristic for Solving the Multichoice Multidimensional Knapsack Problem
A new heuristic for solving the multichoice multidimensional knapsack problem (MMKP) is presented in this paper. The MMKP is first reduced to a multidimensional knapsack problem (MKP). A linear programming relaxation of the resulting MKP is solved, and ...
A fast and scalable multidimensional multiple-choice knapsack heuristic
Special Section on Networks on Chip: Architecture, Tools, and MethodologiesMany combinatorial optimization problems in the embedded systems and design automation domains involve decision making in multidimensional spaces. The multidimensional multiple-choice knapsack problem (MMKP) is among the most challenging of the ...
An exact algorithm for the knapsack sharing problem
In this paper, we develop an exact algorithm for solving the knapsack sharing problem. The algorithm is a new version of the method proposed in Hifi and Sadfi (J. Combin. Optim. 6 (2002) 35). It seems quite efficient in the sense that it solves quickly ...








Comments