TY - GEN

T1 - Wait-free test-and-set

AU - Afek, Yehuda

AU - Gafni, Eli

AU - Tromp, John

AU - Vitanyi, Paul M.B.

N1 - Publisher Copyright:
© 1992, Springer Verlag. All rights reserved.

PY - 1992

Y1 - 1992

N2 - This paper presents an economical, randomized, wait-free construction of an n-process test-and-set bit from read write registers. The test-and-set shared object has two atomic operations, test&set, which atomically reads the bit and sets its value to 1, and the reset operation that resets the bit to 0. We identify two new complexity measures by which to evaluate waitfree algorithms: (a) The amount of randomness used, and (b) ‘Parallel-Time’—the maximum sequential depth of an execution (i.e. longest chain of operations that must precede each other). The previously best known algorithm for n-process test-and-set [Her91] takes an expected Ω(n2) parallel time, and Ω(n4) sequential time per operation, and Ω(n2log n) space per processor. In contrast, our direct implementation improves this on all counts by using O(log n) coin flips, O(log n) parallel time, O(n) sequential time, per operation, and O(n) space per processor. Thus the question on the difference in the expected complexity of randomized constructions of concurrent objects from read/write registers is raised.

AB - This paper presents an economical, randomized, wait-free construction of an n-process test-and-set bit from read write registers. The test-and-set shared object has two atomic operations, test&set, which atomically reads the bit and sets its value to 1, and the reset operation that resets the bit to 0. We identify two new complexity measures by which to evaluate waitfree algorithms: (a) The amount of randomness used, and (b) ‘Parallel-Time’—the maximum sequential depth of an execution (i.e. longest chain of operations that must precede each other). The previously best known algorithm for n-process test-and-set [Her91] takes an expected Ω(n2) parallel time, and Ω(n4) sequential time per operation, and Ω(n2log n) space per processor. In contrast, our direct implementation improves this on all counts by using O(log n) coin flips, O(log n) parallel time, O(n) sequential time, per operation, and O(n) space per processor. Thus the question on the difference in the expected complexity of randomized constructions of concurrent objects from read/write registers is raised.

UR - http://www.scopus.com/inward/record.url?scp=84976670987&partnerID=8YFLogxK

U2 - 10.1007/3-540-56188-9_6

DO - 10.1007/3-540-56188-9_6

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84976670987

SN - 9783540561880

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 85

EP - 94

BT - Distributed Algorithms - 6th International Workshop, WDAG 1992, Proceedings

A2 - Segall, Adrian

A2 - Zaks, Shmuel

PB - Springer Verlag

T2 - 6th International Workshop on Distributed Algorithms, WDAG 1992

Y2 - 2 November 1992 through 4 November 1992

ER -