TY - GEN
T1 - Perfect Matching with Few Link Activations
AU - Mirault, Hugo
AU - Robinson, Peter
AU - Tan, Ming Ming
AU - Zhu, Xianbin
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - We consider the problem of computing a perfect matching problem in a synchronous distributed network, where the network topology corresponds to a complete bipartite graph. The communication between nodes is restricted to activating communication links, which means that instead of sending messages containing a number of bits, each node can only send a pulse over some of its incident links in each round. In the port numbering model, where nodes are unaware of their neighbor’s IDs, we give a randomized algorithm that terminates in Ologn rounds and has a pulse complexity of Onlogn, which corresponds to the number of pulses sent over all links. We also show that randomness is crucial in the port numbering model, as any deterministic algorithm must send at least Ωn2 messages in the standard LOCAL model, where the messages can be of unbounded size. Then, we turn our attention to the KT1 assumption, where each node starts out knowing its neighbors’ IDs. We show that this additional knowledge enables significantly improved bounds even for deterministic algorithms. First, we give an Ologn time deterministic algorithm that sends only On pulses. Finally, we apply this algorithm recursively to obtain an exponential reduction in the time complexity to Olog∗nloglogn, while slightly increasing the pulse complexity to Onlog∗n. All our bounds also hold in the standard CONGEST model with single-bit messages.
AB - We consider the problem of computing a perfect matching problem in a synchronous distributed network, where the network topology corresponds to a complete bipartite graph. The communication between nodes is restricted to activating communication links, which means that instead of sending messages containing a number of bits, each node can only send a pulse over some of its incident links in each round. In the port numbering model, where nodes are unaware of their neighbor’s IDs, we give a randomized algorithm that terminates in Ologn rounds and has a pulse complexity of Onlogn, which corresponds to the number of pulses sent over all links. We also show that randomness is crucial in the port numbering model, as any deterministic algorithm must send at least Ωn2 messages in the standard LOCAL model, where the messages can be of unbounded size. Then, we turn our attention to the KT1 assumption, where each node starts out knowing its neighbors’ IDs. We show that this additional knowledge enables significantly improved bounds even for deterministic algorithms. First, we give an Ologn time deterministic algorithm that sends only On pulses. Finally, we apply this algorithm recursively to obtain an exponential reduction in the time complexity to Olog∗nloglogn, while slightly increasing the pulse complexity to Onlog∗n. All our bounds also hold in the standard CONGEST model with single-bit messages.
KW - distributed graph algorithm
KW - perfect matching
UR - https://www.scopus.com/pages/publications/105008412717
UR - https://www.scopus.com/pages/publications/105008412717#tab=citedBy
U2 - 10.1007/978-3-031-91736-3_28
DO - 10.1007/978-3-031-91736-3_28
M3 - Conference contribution
AN - SCOPUS:105008412717
SN - 9783031917356
T3 - Lecture Notes in Computer Science
SP - 437
EP - 443
BT - Structural Information and Communication Complexity - 32nd International Colloquium, SIROCCO 2025, Proceedings
A2 - Schmid, Ulrich
A2 - Kuznets, Roman
PB - Springer Science and Business Media Deutschland GmbH
T2 - 32nd International Colloquium on Structural Information and Communication Complexity, SIROCCO 2025
Y2 - 2 June 2025 through 4 June 2025
ER -