Improved Bounds for Distributed Load Balancing

Authors Sepehr Assadi, Aaron Bernstein, Zachary Langley

Thumbnail PDF


  • Filesize: 0.54 MB
  • 15 pages

Document Identifiers

Author Details

Sepehr Assadi
  • Rutgers University, New Brunswick, NJ, USA
Aaron Bernstein
  • Rutgers University, New Brunswick, NJ, USA
Zachary Langley
  • Rutgers University, New Brunswick, NJ, USA

Cite AsGet BibTex

Sepehr Assadi, Aaron Bernstein, and Zachary Langley. Improved Bounds for Distributed Load Balancing. In 34th International Symposium on Distributed Computing (DISC 2020). Leibniz International Proceedings in Informatics (LIPIcs), Volume 179, pp. 1:1-1:15, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2020)


In the load balancing problem, the input is an n-vertex bipartite graph G = (C ∪ S, E) - where the two sides of the bipartite graph are referred to as the clients and the servers - and a positive weight for each client c ∈ C. The algorithm must assign each client c ∈ C to an adjacent server s ∈ S. The load of a server is then the weighted sum of all the clients assigned to it. The goal is to compute an assignment that minimizes some function of the server loads, typically either the maximum server load (i.e., the 𝓁_∞-norm) or the 𝓁_p-norm of the server loads. This problem has a variety of applications and has been widely studied under several different names, including: scheduling with restricted assignment, semi-matching, and distributed backup placement. We study load balancing in the distributed setting. There are two existing results in the CONGEST model. Czygrinow et al. [DISC 2012] showed a 2-approximation for unweighted clients with round-complexity O(Δ⁵), where Δ is the maximum degree of the input graph. Halldórsson et al. [SPAA 2015] showed an O(log n / log log n)-approximation for unweighted clients and O(log²n/log log n)-approximation for weighted clients with round-complexity polylog(n). In this paper, we show the first distributed algorithms to compute an O(1)-approximation to the load balancing problem in polylog(n) rounds: - In the CONGEST model, we give an O(1)-approximation algorithm in polylog(n) rounds for unweighted clients. For weighted clients, the approximation ratio is O(log{n}). - In the less constrained LOCAL model, we give an O(1)-approximation algorithm for weighted clients in polylog(n) rounds. Our approach also has implications for the standard sequential setting in which we obtain the first O(1)-approximation for this problem that runs in near-linear time. A 2-approximation is already known, but it requires solving a linear program and is hence much slower. Finally, we note that all of our results simultaneously approximate all 𝓁_p-norms, including the 𝓁_∞-norm.

Subject Classification

ACM Subject Classification
  • Theory of computation → Distributed algorithms
  • Load Balancing
  • Distributed Algorithms
  • Matching
  • Semi-Matching


  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    PDF Downloads


  1. Sepehr Assadi, Aaron Bernstein, and Zachary Langley. Improved bounds for distributed load balancing, 2020. arXiv:2008.04148. URL:
  2. Yossi Azar, Leah Epstein, Yossi Richter, and Gerhard J. Woeginger. All-norm approximation algorithms. J. Algorithms, 52(2):120-133, 2004. URL:
  3. Leonid Barenboim and Gal Oren. Distributed backup placement in one round and its applications to maximum matching and self-stabilization. In Proc. 3rd Symposium on Simplicity in Algorithms (SOSA 2020), pages 99-105, 2020. URL:
  4. Aaron Bernstein, Jacob Holm, and Eva Rotenberg. Online bipartite matching with amortized O(log² n) replacements. J. ACM, 66(5):Art. 37, 23, 2019. URL:
  5. Aaron Bernstein, Tsvi Kopelowitz, Seth Pettie, Ely Porat, and Clifford Stein. Simultaneously load balancing for every p-norm, with reassignments. In Proc. 8th Innovations in Theoretical Computer Science Conference (ITCS 2017), volume 67 of LIPIcs. Leibniz Int. Proc. Inform., pages Art. No. 51, 14. Schloss Dagstuhl. Leibniz-Zent. Inform., Wadern, 2017. URL:
  6. J. Bruno, E. G. Coffman, Jr., and R. Sethi. Scheduling independent tasks to reduce mean finishing time. Comm. ACM, 17:382-387, 1974. URL:
  7. Deeparnab Chakrabarty and Chaitanya Swamy. Simpler and better algorithms for minimum-norm load balancing. In Proc. 27th Annual European Symposium on Algorithms (ESA 2019), volume 144 of LIPIcs. Leibniz Int. Proc. Inform., pages Art. No. 27, 12. Schloss Dagstuhl. Leibniz-Zent. Inform., Wadern, 2019. URL:
  8. Andrzej Czygrinow, Michal Hanćkowiak, Edyta Szymańska, and Wojciech Wawrzyniak. Distributed 2-approximation algorithm for the semi-matching problem. In Proc. 26th International Symposium on Distributed Computing (DISC 2012), volume 7611 of LNCS, pages 210-222. Springer, Heidelberg, 2012. URL:
  9. Jittat Fakcharoenphol, Bundit Laekhanukit, and Danupon Nanongkai. Faster algorithms for semi-matching problems. ACM Trans. Algorithms, 10(3):Art. 14, 23, 2014. URL:
  10. Martin Gairing, Thomas Lücking, Marios Mavronicolas, and Burkhard Monien. The price of anarchy for restricted parallel links. Parallel Process. Lett., 16(1):117-131, 2006. URL:
  11. Anupam Gupta, Amit Kumar, and Cliff Stein. Maintaining assignments online: matching, scheduling, and flows. In Proc. 25th Annual ACM-SIAM Symposium on Discrete Algorithms (SODA 2014), pages 468-479. ACM, 2014. URL:
  12. Magnús M. Halldórsson, Sven Köhler, Boaz Patt-Shamir, and Dror Rawitz. Distributed backup placement in networks. Distrib. Comput., 31(2):83-98, 2018. URL:
  13. Nicholas J. A. Harvey, Richard E. Ladner, László Lovász, and Tami Tamir. Semi-matchings for bipartite graphs and load balancing. J. Algorithms, 59(1):53-78, 2006. URL:
  14. W. A. Horn. Minimizing average flow time with parallel machines. Oper. Res., 21(3), 1973. URL:
  15. Donggu Kang and James Payor. Flow rounding, 2015. arXiv:1507.08139. URL:
  16. Sven Köhler, Volker Turau, and Gerhard Mentges. Self-stabilizing local k-placement of replicas with minimal variance. In Proc. 14th Stabilization, Safety, and Security of Distributed Systems (SSS 2012), pages 16-30, 2012. URL:
  17. Christian Konrad and Adi Rosén. Approximating semi-matchings in streaming and in two-party communication. ACM Trans. Algorithms, 12(3):Art. 32, 21, 2016. URL:
  18. Anshul Kothari, Subhash Suri, Csaba D. Tóth, and Yunhong Zhou. Congestion games, load balancing, and price of anarchy. In Proc. 1st Combinatorial and Algorithmic Aspects of Networking, volume 3405 of LNCS, pages 13-27. Springer, Berlin, 2005. URL:
  19. Yixun Lin and Wenhua Li. Parallel machine scheduling of machine-dependent jobs with unit-length. European J. Oper. Res., 156(1):261-266, 2004. URL:
  20. Zvi Lotker, Boaz Patt-Shamir, and Seth Pettie. Improved distributed approximate matching. J. ACM, 62(5):Art. 38, 17, 2015. URL:
  21. Renita Machado and Sirin Tekinay. A survey of game-theoretic approaches in wireless sensor networks. Comput. Networks, 52(16):3047-3061, 2008. URL:
  22. Gal Oren, Leonid Barenboim, and Harel Levin. Distributed fault-tolerant backup-placement in overloaded wireless sensor networks. In Proc. 9th International Conference on Broadband Communications, Networks, and Systems (BROADNETS 2018), pages 212-224, 2018. URL:
  23. Narayanan Sadagopan, Mitali Singh, and Bhaskar Krishnamachari. Decentralized utility-based sensor network design. Mobile Networks and Applications, 11(3):341-350, 2006. URL:
  24. Daniel D. Sleator and Robert E. Tarjan. A data structure for dynamic trees. J. Comput. Syst. Sci., 26(3):362-391, 1983. URL:
  25. Subhash Suri, Csaba D. Tóth, and Yunhong Zhou. Uncoordinated load balancing and congestion games in P2P systems. In Proc. 3rd International Workshop on Peer-to-Peer Systems (IPTPS 2004), pages 123-130, 2004. URL:
  26. Subhash Suri, Csaba D. Tóth, and Yunhong Zhou. Selfish load balancing and atomic congestion games. Algorithmica, 47(1):79-96, 2007. URL:
Questions / Remarks / Feedback

Feedback for Dagstuhl Publishing

Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail