Shortest Paths in a Hybrid Network Model

We introduce a communication model for hybrid networks, where nodes have access to two different communication modes: a local mode where communication is only possible between specific pairs of nodes, and a global mode where communication between any pair of nodes is possible. This can be motivated, for instance, by wireless networks in which we combine direct device-to-device communication (e.g., using WiFi) with communication via a shared infrastructure (like base stations, the cellular network, or satellites). Typically, communication over short-range connections is cheaper and can be done at a much higher rate. Hence, we are focusing here on the LOCAL model (in which the nodes can exchange an unbounded amount of information in each round) for the local connections while for the global communication we assume the so-called node-capacitated clique model, where in each round every node can exchange only $O(\log n)$-bit messages with just $O(\log n)$ other nodes. In order to explore the power of combining local and global communication, we study the impact of hybrid communication on the complexity of computing shortest paths in the graph given by the local connections. Specifically, for the all-pairs shortest paths problem (APSP), we show that an exact solution can be computed in time $\tilde O\big(n^{2/3}\big)$ and that approximate solutions can be computed in time $\tilde \Theta\big(\!\sqrt{n}\big)$. For the single-source shortest paths problem (SSSP), we show that an exact solution can be computed in time $\tilde O\big(\!\sqrt{\mathsf{SPD}}\big)$, where $\mathsf{SPD}$ denotes the shortest path diameter. We further show that a $\big(1\!+\!o(1)\big)$-approximate solution can be computed in time $\tilde O\big(n^{1/3}\big)$. Additionally, we show that for every constant $\varepsilon>0$, it is possible to compute an $O(1)$-approximate solution in time $\tilde O(n^\varepsilon)$.

[1]  Michael Elkin,et al.  Distributed exact shortest paths in sublinear time , 2017, STOC.

[2]  Mohsen Ghaffari,et al.  Near-Optimal Scheduling of Distributed Algorithms , 2015, PODC.

[3]  Aravind Srinivasan,et al.  Chernoff-Hoeffding bounds for applications with limited independence , 1995, SODA '93.

[4]  Günter Schäfer,et al.  A survey on automatic configuration of virtual private networks , 2011, Comput. Networks.

[5]  Boaz Patt-Shamir,et al.  Minimum-Weight Spanning Tree Construction in O(log log n) Communication Rounds , 2005, SIAM J. Comput..

[6]  Stefan Schmid,et al.  Characterizing the algorithmic complexity of reconfigurable data center architectures , 2018, ANCS.

[7]  Christian Scheideler,et al.  Competitive Routing in Hybrid Communication Networks , 2017, ALGOSENSORS.

[8]  Christoph Lenzen,et al.  Optimal deterministic routing and sorting on the congested clique , 2012, PODC '13.

[9]  Keren Censor-Hillel,et al.  Fast approximate shortest paths in the congested clique , 2019, Distributed Computing.

[10]  Danupon Nanongkai,et al.  A Faster Distributed Single-Source Shortest Paths Algorithm , 2017, 2018 IEEE 59th Annual Symposium on Foundations of Computer Science (FOCS).

[11]  Sandeep Sen,et al.  A simple and linear time randomized algorithm for computing sparse spanners in weighted graphs , 2007 .

[12]  Christoph Lenzen,et al.  Near-Optimal Approximate Shortest Paths and Transshipment in Distributed and Streaming Models , 2016, DISC.

[13]  Christoph Lenzen,et al.  Efficient distributed source detection with limited bandwidth , 2013, PODC '13.

[14]  Tomasz Jurdzinski,et al.  MST in O(1) Rounds of Congested Clique , 2018, SODA.

[15]  Danupon Nanongkai,et al.  Distributed approximation algorithms for weighted shortest paths , 2014, STOC.

[16]  Paramvir Bahl,et al.  Augmenting data center networks with multi-gigabit wireless links , 2011, SIGCOMM.

[17]  Mihalis Yannakakis,et al.  High-Probability Parallel Transitive-Closure Algorithms , 1991, SIAM J. Comput..

[18]  Jamal N. Al-Karaki,et al.  Routing protocols in wireless mesh networks: a survey , 2017, Int. J. Inf. Commun. Technol..

[19]  Matteo Pontecorvi,et al.  A Deterministic Distributed Algorithm for Exact Weighted All-Pairs Shortest Paths in Õ(n 3/2 ) Rounds , 2018, PODC.

[20]  Thatchaphol Saranurak,et al.  Distributed Exact Weighted All-Pairs Shortest Paths in Õ(n^{5/4}) Rounds , 2017, 2017 IEEE 58th Annual Symposium on Foundations of Computer Science (FOCS).

[21]  Christian Scheideler,et al.  Distributed Computation in Node-Capacitated Networks , 2019, SPAA.

[22]  Keren Censor-Hillel,et al.  Near-Linear Lower Bounds for Distributed Distance Computations, Even in Sparse Networks , 2016, DISC.

[23]  Amin Vahdat,et al.  Helios: a hybrid electrical/optical switch architecture for modular data centers , 2010, SIGCOMM '10.

[24]  Guy E. Blelloch,et al.  Near linear-work parallel SDD solvers, low-diameter decomposition, and low-stretch subgraphs , 2011, SPAA '11.

[25]  Sandeep Sen,et al.  A simple and linear time randomized algorithm for computing sparse spanners in weighted graphs , 2007, Random Struct. Algorithms.

[26]  Debarshi Kumar Sanyal,et al.  An overview of device-to-device communication in cellular networks , 2017, ICT Express.

[27]  Christian Scheideler,et al.  Distributed Monitoring of Network Properties: The Power of Hybrid Networks , 2017, ICALP.

[28]  Fabian Kuhn,et al.  On the power of the congested clique model , 2014, PODC.

[29]  Danupon Nanongkai,et al.  Distributed exact weighted all-pairs shortest paths in near-linear time , 2018, STOC.

[30]  Boaz Patt-Shamir,et al.  Fast Partial Distance Estimation and Applications , 2014, PODC.

[31]  Anis Tell,et al.  SD-WAN: A Modern Hybrid-WAN to Enable Digital Transformation for Businesses , 2018 .

[32]  Mohsen Ghaffari,et al.  Improved distributed algorithms for exact shortest paths , 2018, STOC.

[33]  Roger Wattenhofer,et al.  Optimal distributed all pairs shortest paths and applications , 2012, PODC '12.

[34]  Christoph Lenzen,et al.  Algebraic methods in the congested clique , 2015, Distributed Computing.

[35]  Roger Wattenhofer,et al.  Networks cannot compute their diameter in sublinear time , 2012, SODA.

[36]  Boaz Patt-Shamir,et al.  Fast Routing Table Construction Using Small Messages , 2012, ArXiv.

[37]  David Peleg,et al.  Distributed Algorithms for Network Diameter and Girth , 2012, ICALP.