Dynamic Balanced Graph Partitioning

This paper initiates the study of the classic balanced graph partitioning problem from an online perspective: Given an arbitrary sequence of pairwise communication requests between $n$ nodes, with patterns that may change over time, the objective is to service these requests efficiently by partitioning the nodes into $\ell$ clusters, each of size $k$, such that frequently communicating nodes are located in the same cluster. The partitioning can be updated dynamically by migrating nodes between clusters. The goal is to devise online algorithms which jointly minimize the amount of inter-cluster communication and migration cost. The problem features interesting connections to other well-known online problems. For example, scenarios with $\ell=2$ generalize online paging, and scenarios with $k=2$ constitute a novel online variant of maximum matching. We present several lower bounds and algorithms for settings both with and without cluster-size augmentation. In particular, we prove that any deterministic online algorithm has a competitive ratio of at least $k$, even with significant augmentation. Our main algorithmic contributions are an $O(k \log{k})$-competitive deterministic algorithm for the general setting with constant augmentation, and a constant competitive algorithm for the maximum matching variant.

[1]  Vijay V. Vazirani,et al.  Finding k-cuts within twice the optimal , 1991, [1991] Proceedings 32nd Annual Symposium of Foundations of Computer Science.

[2]  Seeun Umboh Online Network Design Algorithms via Hierarchical Decompositions , 2015, SODA.

[3]  Yossi Azar,et al.  On-line generalized Steiner problem , 1996, SODA '96.

[4]  Mário César San Felice A Randomized O ( log n )-Competitive Algorithm for the Online Connected Facility Location Problem , 2014 .

[5]  Stefan Schmid,et al.  Competitive Clustering of Stochastic Communication Patterns on a Ring , 2017, NETYS.

[6]  David P. Williamson,et al.  A Randomized $$\mathrm {O}(\log n)$$O(logn)-Competitive Algorithm for the Online Connected Facility Location Problem , 2016, Algorithmica.

[7]  Piotr Indyk,et al.  On page migration and other relaxed task systems , 1997, SODA '97.

[8]  Joseph Naor,et al.  Divide-and-conquer approximation algorithms via spreading metrics , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[9]  Boaz Patt-Shamir,et al.  Rent, Lease, or Buy: Randomized Algorithms for Multislope Ski Rental , 2012, SIAM J. Discret. Math..

[10]  Robert Krauthgamer,et al.  A polylogarithmic approximation of the minimum bisection , 2000, Proceedings 41st Annual Symposium on Foundations of Computer Science.

[11]  Fillia Makedon,et al.  Approximation algorithms for VLSI partition problems , 1990, IEEE International Symposium on Circuits and Systems.

[12]  Marek Karpinski,et al.  Polynomial Time Approximation Schemes for Dense Instances of NP-Hard Problems , 1999, J. Comput. Syst. Sci..

[13]  Leah Epstein,et al.  On Variants of File Caching , 2011, ICALP.

[14]  Harald Räcke,et al.  Optimal hierarchical decompositions for congestion minimization in networks , 2008, STOC.

[15]  Lucian Popa,et al.  What we talk about when we talk about cloud network performance , 2012, CCRV.

[16]  Lyle A. McGeoch,et al.  A strongly competitive randomized paging algorithm , 1991, Algorithmica.

[17]  Michael I. Jordan,et al.  Managing data transfers in computer clusters with orchestra , 2011, SIGCOMM.

[18]  Luca Foschini,et al.  Balanced Partitions of Trees and Applications , 2012, Algorithmica.

[19]  Satish Rao,et al.  Expander flows, geometric embeddings and graph partitioning , 2004, STOC '04.

[20]  Amos Fiat,et al.  Distributed paging for general networks , 1996, SODA '96.

[21]  Stefan Schmid,et al.  Efficient Distributed Workload (Re-)Embedding , 2019, SIGMETRICS.

[22]  Robert E. Tarjan,et al.  Amortized efficiency of list update and paging rules , 1985, CACM.

[23]  Amos Fiat,et al.  Competitive distributed file allocation , 1993, STOC '93.

[24]  Leah Epstein,et al.  Online File Caching with Rejection Penalties , 2015, Algorithmica.

[25]  Jaroslaw Byrka,et al.  Dynamic beats fixed: on phase-based algorithms for file migration , 2017, ICALP.

[26]  Vijay V. Vazirani,et al.  Finding k Cuts within Twice the Optimal , 1995, SIAM J. Comput..

[27]  Yuval Rabani,et al.  Competitive Algorithms for Distributed Data Management , 1995, J. Comput. Syst. Sci..

[28]  Sandy Irani,et al.  Page replacement with multi-size pages and applications to Web caching , 1997, STOC '97.

[29]  Joseph Naor,et al.  Fast approximate graph partitioning algorithms , 1997, SODA '97.

[30]  Anna R. Karlin,et al.  Competitive randomized algorithms for non-uniform problems , 1990, SODA '90.

[31]  Marek Chrobak,et al.  Competitive analysis of randomized paging algorithms , 2000, Theor. Comput. Sci..

[32]  Konstantin Andreev,et al.  Balanced Graph Partitioning , 2004, SPAA '04.

[33]  Amos Fiat,et al.  Competitive Paging Algorithms , 1991, J. Algorithms.

[34]  Robert Krauthgamer,et al.  Partitioning graphs into balanced components , 2009, SODA.

[35]  Shang-Hua Teng,et al.  How Good is Recursive Bisection? , 1997, SIAM J. Sci. Comput..

[36]  Stefan Schmid,et al.  Online Balanced Repartitioning , 2015, DISC.

[37]  Robert Krauthgamer,et al.  Approximating the minimum bisection size (extended abstract) , 2000, STOC '00.

[38]  Sandy Irani,et al.  Page Replacement with Multi-Size Pages and Applications to Web Caching , 2002, Algorithmica.

[39]  David S. Johnson,et al.  Some Simplified NP-Complete Graph Problems , 1976, Theor. Comput. Sci..