Load balancing strategies for symbolic vision computations

Most intermediate and high-level vision algorithms manipulate symbolic features. A key operation in these vision algorithms is to search symbolic features satisfying certain geometric constraints. Parallelizing this symbolic search needs a non-trivial algorithmic technique due to the unpredictable workload. In this paper, we propose load balancing strategies for parallelizing symbolic search operations on distributed memory machines. By using an initial workload estimate, we first partition the computations such that the workload is distributed evenly across the processors. In addition, we perform fast migrations dynamically to adapt to the evolving workload. To demonstrate the usefulness of our load balancing strategies, experiments were conducted on an IBM SP2 and a Cray T3D. Our results show that our task migration strategy can balance the unpredictable workload with little overhead. Our code using C and MPI is portable onto other high performance computing platforms.

[1]  K. Ramesh Babu,et al.  Linear Feature Extraction and Description , 1979, IJCAI.

[2]  Katherine Yelick,et al.  Runtime Support for Portable Distributed Data Structures , 1995, LCR.

[3]  Alok N. Choudhary,et al.  An Efficient Heuristic Scheme for Dynamic Remapping of Parallel Computations , 1993, Parallel Comput..

[4]  David A. Bader,et al.  Practical parallel algorithms for dynamic data redistribution, median finding, and selection , 1995, Proceedings of International Conference on Parallel Processing.

[5]  Bradley C. Kuszmaul,et al.  Cilk: an efficient multithreaded runtime system , 1995, PPOPP '95.

[6]  Ramakant Nevatia,et al.  Matching Images Using Linear Features , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Srinivas S. Ravela.  A Survey of Reactivity , 1992 .

[8]  Yongwha Chung,et al.  A fast asynchronous algorithm for linear feature extraction on IBM SP-2 , 1995, Proceedings of Conference on Computer Architectures for Machine Perception.

[9]  Geoffrey C. Fox,et al.  Performance evaluation of load balancing algorithms for parallel single-phase iterative PDE solvers , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.