Distributionally Linearizable Data Structures

Alistarh, Dan; Brown, Thomas J.; Kopinsky, Justin; Li, Jerry Z.; Nadiradze, Giorgi

doi:10.48550/arxiv.1804.01018

Cited by 1 publication

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Seminal work by Azar, Broder, Karlin, and Upfal [3] showed that, if we place n unit weights into n bins by the d-choice process with d ≥ 2, then, surprisingly, the maximum load is reduced to Θ(log log n/ log d). A technical tour-de-force by Berenbrink, Czumaj, Steger, and Vcking [4] extended this result to the "heavily-loaded" case where m n, showing that in this case the maximum load is m/n + log log n/ log d + O (1) with failure probability at most 1/ poly n. An elegant alternative proof for a slightly weaker version of this result was later provided by Talwar and Wieder [13].…”

Section: Introductionmentioning

confidence: 90%

“…We consider balls-into-bins processes where a sequence of m weights are placed into n bins via some randomized procedure, with the goal of minimizing the load imbalance between the most loaded and the least loaded bin. This family of processes has been used to model several practical allocation problems, such as load-balancing [3,7,11], hashing [5], or even relaxed data structures [2,1].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Averaging Load Balancing on Cycles

Alistarh¹,

Nadiradze²,

Sabour³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

We consider the following dynamic load-balancing process: given an underlying graph G with n nodes, in each step t ≥ 0, one unit of load is created, and placed at a randomly chosen graph node. In the same step, the chosen node picks a random neighbor, and the two nodes balance their loads by averaging them. We are interested in the expected gap between the minimum and maximum loads at nodes as the process progresses, and its dependence on n and on the graph structure.Similar variants of the above graphical balanced allocation process have been studied by Peres, Talwar, and Wieder [10], and by Sauerwald and Sun [12] for regular graphs. These authors left as open the question of characterizing the gap in the case of cycle graphs in the dynamic case, where weights are created during the algorithm's execution. For this case, the only known upper bound is of O(n log n), following from a majorization argument due to [10], which analyzes a related graphical allocation process.In this paper, we provide an upper bound of O( √ n log n) on the expected gap of the above process for cycles of length n. We introduce a new potential analysis technique, which enables us to bound the difference in load between k-hop neighbors on the cycle, for any k ≤ n/2. We complement this with a "gap covering" argument, which bounds the maximum value of the gap by bounding its value across all possible subsets of a certain structure, and recursively bounding the gaps within each subset. We provide analytical and experimental evidence that our upper bound on the gap is tight up to a logarithmic factor.

show abstract