TY - JOUR
T1 - Implementing three exchange read operations for distributed atomic storage
AU - Georgiou, Chryssis
AU - Hadjistasi, Theophanis
AU - Nicolaou, Nicolas
AU - Schwarzmann, Alexander A.
N1 - DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.
PY - 2022/5
Y1 - 2022/5
N2 - Communication latency typically dominates the performance of message-passing systems, and consequently defines the efficiency of operations of algorithms implementing atomic read/write objects in asynchronous, crash-prone, message-passing systems. Here latency is measured in terms of the number of communication exchanges (or simply exchanges) involved in each operation. We present four algorithms, two for the single-writer/multiple-reader (SWMR) and two for the multi-writer/multiple-reader (MWMR) settings, that allow reads to take two or three exchanges, advancing the state-of-the-art in this area. Writes take the same number of exchanges as in prior works (i.e., two for SWMR and four for MWMR settings). In contrast with existing efficient implementations, ours come with no constraints on reader participation in both settings, and on the number of writers in the MWMR setting. Correctness of algorithms is rigorously argued. We conclude with an empirical study demonstrating the practicality of the algorithms, and identifying settings in which their read performance, is clearly superior compared to relevant algorithms.
AB - Communication latency typically dominates the performance of message-passing systems, and consequently defines the efficiency of operations of algorithms implementing atomic read/write objects in asynchronous, crash-prone, message-passing systems. Here latency is measured in terms of the number of communication exchanges (or simply exchanges) involved in each operation. We present four algorithms, two for the single-writer/multiple-reader (SWMR) and two for the multi-writer/multiple-reader (MWMR) settings, that allow reads to take two or three exchanges, advancing the state-of-the-art in this area. Writes take the same number of exchanges as in prior works (i.e., two for SWMR and four for MWMR settings). In contrast with existing efficient implementations, ours come with no constraints on reader participation in both settings, and on the number of writers in the MWMR setting. Correctness of algorithms is rigorously argued. We conclude with an empirical study demonstrating the practicality of the algorithms, and identifying settings in which their read performance, is clearly superior compared to relevant algorithms.
KW - Atomicity
KW - Fault-tolerance
KW - Read/write objects
KW - Shared memory
UR - http://www.scopus.com/inward/record.url?scp=85124270427&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124270427&partnerID=8YFLogxK
U2 - 10.1016/j.jpdc.2022.01.024
DO - 10.1016/j.jpdc.2022.01.024
M3 - Article
AN - SCOPUS:85124270427
SN - 0743-7315
VL - 163
SP - 97
EP - 113
JO - Journal of Parallel and Distributed Computing
JF - Journal of Parallel and Distributed Computing
ER -