TY - JOUR
T1 - Component diagnosability in terms of component connectivity of hypercube-based compound networks
AU - Liu, Jiafei
AU - Zhou, Shuming
AU - Wang, Dajin
AU - Zhang, Hong
N1 - Funding Information:
The authors would like to express their sincere gratitude to all reviewers for valuable suggestions, which are helpful in improving and clarifying the original manuscript. This work was partly supported by the National Natural Science Foundation of China (Nos. 61977016 and 61572010 ), Natural Science Foundation of Fujian Province (Nos. 2020J01164 , 2017J01738 ).
Publisher Copyright:
© 2022 Elsevier Inc.
PY - 2022/4
Y1 - 2022/4
N2 - Enhancing the invulnerability of multiprocessor systems against malicious attacks has been regarded as one of the important issues in network science and big data era. Thus, in order to firmly characterize the robustness of systems, several variants of classic connectivity have been proposed so far. The component connectivity is a significant metric in evaluating the robustness and fault tolerability of interconnection network. For an interconnection network G and a positive integer h, the (h+1)-component connectivity of G, denoted cκh+1(G), is the cardinality of a minimum vertex cut F such that G−F has at least h+1 connected components. Based on component connectivity, component diagnosability has been proposed to measure the self-diagnosis capability of multiprocessor systems. In this paper, we suggest some characterizations of the (h+1)-component connectivity of a class of regular networks under some restrictions. Furthermore, we establish the relationship between component connectivity and component diagnosability of one class of networks. As by-products, we present the (h+1)-component diagnosability of the state-of-the-art compound networks based on hypercube, such as bicube network, generalized exchanged hypercube, hierarchical hypercube, half-hypercube, and so on.
AB - Enhancing the invulnerability of multiprocessor systems against malicious attacks has been regarded as one of the important issues in network science and big data era. Thus, in order to firmly characterize the robustness of systems, several variants of classic connectivity have been proposed so far. The component connectivity is a significant metric in evaluating the robustness and fault tolerability of interconnection network. For an interconnection network G and a positive integer h, the (h+1)-component connectivity of G, denoted cκh+1(G), is the cardinality of a minimum vertex cut F such that G−F has at least h+1 connected components. Based on component connectivity, component diagnosability has been proposed to measure the self-diagnosis capability of multiprocessor systems. In this paper, we suggest some characterizations of the (h+1)-component connectivity of a class of regular networks under some restrictions. Furthermore, we establish the relationship between component connectivity and component diagnosability of one class of networks. As by-products, we present the (h+1)-component diagnosability of the state-of-the-art compound networks based on hypercube, such as bicube network, generalized exchanged hypercube, hierarchical hypercube, half-hypercube, and so on.
KW - Component connectivity
KW - Component diagnosability
KW - Multiprocessor systems
KW - Robustness
UR - http://www.scopus.com/inward/record.url?scp=85122525064&partnerID=8YFLogxK
U2 - 10.1016/j.jpdc.2021.12.004
DO - 10.1016/j.jpdc.2021.12.004
M3 - Article
AN - SCOPUS:85122525064
SN - 0743-7315
VL - 162
SP - 17
EP - 26
JO - Journal of Parallel and Distributed Computing
JF - Journal of Parallel and Distributed Computing
ER -