Simulation results showed that, for all traffic profiles, the new routing algorithms designed for the topologies of circulant networks with rectangular L-shapes outperform other algorithms with similar metrics in terms of memory consumption.
Abstract
This article examines series of families of two-dimensional circulant networks with rectangular L-shapes, optimal in diameter, as network-on-chip topologies with a minimal number of crossings between the links and a bounded length of the maximum link that does not depend on the network size. New network-on-chip routing algorithms, which use the coordinates of three adjacent zeros in the planar graph embedding to calculate the shortest paths, were investigated and simulated. A key advantage of the proposed routing algorithms is that they require minimal input data to calculate the shortest paths. Four routing algorithms were implemented in the Noxim network-on-chip simulator using the optimal graphs of the circulant network families under study: the new analytical algorithm, the traditional Dijkstra algorithm, the routing algorithm with virtual coordinates, and the clockwise routing algorithm. The above algorithms were compared under various traffic profiles across three metrics: average and maximum latencies and network throughput. The simulation results showed that, for all traffic profiles, the new routing algorithms designed for the topologies of circulant networks with rectangular L-shapes outperform other algorithms with similar metrics in terms of memory consumption.
The Single Source Shortest Path (SSSP) problem focuses on finding the shortest paths from a source vertex to all other vertices in a graph. This problem is crucial in various domains, including transportation, telecommunications and urban planning. Although Dijkstra's algorithm is effective for solving the SSSP in general cases, in some special cases, such as two-dimensional mesh networks (2DMN), its performance is not satisfactory. To address this limitation, this paper presents an improved approach, i.e., the row-by-row dynamic programming (RRDP) algorithm, which is specifically optimized for 2DMNs and can significantly reduce computation times, although in worst case the complexity of RRDP is higher than Dijkstra's algorithm. We tested the RRDP algorithm against the traditional Dijkstra's algorithm using the same 2DMN instances, and found that RRDP runs much faster than Dijkstra's algorithm on most instances. This innovation offers new insights and practical solutions for improving path planning algorithms in grid network environments.
S. Sun, Yuwei Sang, Zhang-Hua Fu· 2026 6th International Confe...· 0 citations
This paper identifies an algebraic property of cycles, which is called centripetalism, that characterizes the existence of unique stable routings for all possible destinations in a network and failure scenarios and presents the Consistent-Tree algorithm, which either produces a stable routing or reports the presence of a non-centripetal cycle.
Ricardo Santos, J. L. Sobrinho· Proceedings of the ACM SIGCO...· 0 citations
The results support HON as a simple low-degree construction for structured inter-group communication, whereas higher-radix, adaptive, or more richly connected fabrics remain better suited to less structured traffic and larger bandwidth demand.
Han Ni Soe, Yao Zhang, Zhipeng Xu· Parallel Processing Letters· 0 citations
The recursive match networks (RMNs) represent an important family of communication networks characterized by their regularity and high fault tolerance, including the logic graph of the data center network BCube, the interconnection networks bijective connection (BC) networks, and even potentially other future networks. Protection routing is a key technology to enhance the reliability of communication networks, where dual completely independent spanning trees (dual-CISTs) have garnered significant attention as they suffice to configure a protection routing. In this paper, we propose a novel concept of dual protection routing spanning trees (dual-PRSTs) and demonstrate their application for protection routing construction in RMNs. Compared with dual-CISTs, dual-PRSTs provide more relaxed conditions, allowing for link intersections, internal vertex intersections, and vertices each serving as an inner-vertex in both trees, which are not permitted in dual-CISTs. We show that the research results in this paper are directly applicable to BCube, the BC networks, and other networks that fall within the definition of RMNs. Furthermore, it is proven that as long as a communication network contains dual-PRSTs, they can be utilized to configure a protection routing in it, although the protection routing scheme is conducted in RMNs. Empirical evaluations demonstrate that the protection routing based on dual-PRSTs is not only effective with competitive performance, but also generally outperforms that built with dual-CISTs in terms of path length metrics.
Bai Yin, Qianru Zhou, B. Cheng et al.· IEEE Transactions on Network...· 0 citations
This paper presents a comparative analysis of two variants of a classical algorithm for finding the shortest path in a connected graph. The first variant uses an adjacency matrix (AM) to verify the existence of an edge (arc) between two vertices, while the second variant performs the same verification using an adjacency list (AL). The objective of this study is to examine how graph density affects the performance of the two algorithmic modifications depending on the data structure used. A total of 95 graphs were analyzed, grouped into five sets from 100 to 500 in increments of 100. For each group, 19 graphs were generated with densities ranging from 5% to 95% in increments of 5%. The methodology includes analyzing the number of iterations, assignments, and comparisons executed by the algorithms for all graphs. The initial hypothesis assumed that the total number of operations would always be lower when using an AL instead of an adjacency matrix, regardless of graph density. The results demonstrate that this assumption is incorrect: for densities above 82%, the total number of operations is lower when using an adjacency matrix, whereas the AL is more efficient for densities below 82%, with its efficiency increasing as density decreases. These findings are particularly important for mobile technologies, as they support the design of more efficient pathfinding solutions that optimize performance and energy consumption in mobile applications.
V. Kralev, Radoslava Kraleva, Aleksandra Popova· International Journal of Int...· 0 citations
This work considers the problem of finding, for a given degree sequence, the network structure displaying the smallest possible average shortest-path length and proposes a fast algorithm to construct approximate solutions to such a degree-constrained distance-minimization problem.