Improved Bounds for Distributed Load Balancing

08/10/2020 ∙ by Sepehr Assadi, et al. ∙ 0

In the load balancing problem, the input is an n-vertex bipartite graph G = (C ∪ S, E) 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. 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(Δ^5), where Δ is the maximum degree of the input graph. Halldórsson et al. [SPAA 2015] showed an O(logn/loglogn)-approximation for unweighted clients and O(log^2n/loglogn)-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(logn). 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.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.