Abstract
We examine the use of atomic instructions in implementing barriers that do not require previously initialized memory. We show hown identical processes can use uninitialized shared memory to elect a leader that then initialized the shared memory. The processes first use the uninitialized memory to obtain unique identifiers in the range 0 ton−1 and then meet at a barrier. After passing the barrier, the leader initializes the shared memory. Whenn is not a power of 2 this barrier implementation, a tournament algorithm, avoids extra work by taking advantage of information implicit in the algorithm for obtaining the unique identifiers. The only atomic instruction that we require is one that complements a bit.
Similar content being viewed by others
References
J. L. W. Kessels, Arbitration without Common Modifiable Variables,Acta Informatica,17:135–141 (1982).
G. L. Peterson, Myths about the Mutual Exclusion Problem,Information Processing Letters 12(3):115–116 (1981).
E. D. Brooks, III, The Butterfly Barrier,Intl. J. of Par. Prog. 15(4):295–307 (1986).
D. Hensgen, R. Finkel, U. Manber, Two Algorithms for Barrier Synchronization,Intl. J. of Par. Prog.,17(1):1–17 (1988).
B. K. Szymanski, A Simple Solution to Lamport's Concurrent Programming Problem with Linear Wait,Proc. of the 1988 Intl.Conf. on Supercomputing, St. Malo, pp. 621–626 (1988).
N. S. Arenstorf and H. F. Jordan, Comparing Barrier Algorithms, Technical Report, Computer Systems Design Group, Department of Electrical and Computer Engineering, Univerity of Colorado (1987).
T. S. Axelrod, Effects of Synchronization Barriers on Multiprocessor Performance,Parallel Computing,3:129–140 (1986).
L. Lamport, A Fast Mutual Exclusion Algorithm,ACM Transactions on Computer Systems 5(1):1–11 (1987).
R. A. Finkel, Lightweight Tasks under Unix, Department of Computer Science, University of Kentucky, TR 103-87 (1987).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hemmendinger, D. Initializing memory shared by several processors. Int J Parallel Prog 18, 241–253 (1989). https://doi.org/10.1007/BF01407858
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01407858