HBP: Hotness Balanced Partition for Prioritized Iterative Graph Computations

Author(s):  
Shufeng Gong ◽  
Yanfeng Zhang ◽  
Ge Yu
Keyword(s):  
2021 ◽  
Vol 3 (1) ◽  
Author(s):  
Xianyue Li ◽  
Yufei Pang ◽  
Chenxia Zhao ◽  
Yang Liu ◽  
Qingzhen Dong

AbstractGraph partition is a classical combinatorial optimization and graph theory problem, and it has a lot of applications, such as scientific computing, VLSI design and clustering etc. In this paper, we study the partition problem on large scale directed graphs under a new objective function, a new instance of graph partition problem. We firstly propose the modeling of this problem, then design an algorithm based on multi-level strategy and recursive partition method, and finally do a lot of simulation experiments. The experimental results verify the stability of our algorithm and show that our algorithm has the same good performance as METIS. In addition, our algorithm is better than METIS on unbalanced ratio.


10.37236/252 ◽  
2009 ◽  
Vol 16 (1) ◽  
Author(s):  
Ido Ben-Eliezer ◽  
Michael Krivelevich

For a graph $G=(V,E)$ of even order, a partition $(V_1,V_2)$ of the vertices is said to be perfectly balanced if $|V_1|=|V_2|$ and the numbers of edges in the subgraphs induced by $V_1$ and $V_2$ are equal. For a base graph $H$ define a random graph $G(H,p)$ by turning every non-edge of $H$ into an edge and every edge of $H$ into a non-edge independently with probability $p$. We show that for any constant $\epsilon$ there is a constant $\alpha$, such that for any even $n$ and a graph $H$ on $n$ vertices that satisfies $\Delta(H)-\delta(H) \leq \alpha n$, a graph $G$ distributed according to $G(H,p)$, with ${\epsilon\over n} \leq p \leq 1-{\epsilon\over n}$, admits a perfectly balanced partition with probability exponentially close to $1$. As a direct consequence we get that for every $p$, a random graph from $G(n,p)$ admits a perfectly balanced partition with probability tending to $1$.


2003 ◽  
Vol 13 (04) ◽  
pp. 303-316 ◽  
Author(s):  
MATTIAS ANDERSSON ◽  
JOACHIM GUDMUNDSSON ◽  
CHRISTOS LEVCOPOULOS ◽  
GIRI NARASIMHAN

To better handle situations where additional resources are available to carry out a task, many problems from the manufacturing industry involve dividing a task into a number of smaller tasks, while optimizing a specific objective function. In this paper we consider the problem of partitioning a given set [Formula: see text] of n points in the plane into k subsets, [Formula: see text], such that [Formula: see text] is minimized. Variants of this problem arise in applications from the shipbuilding industry. We show that this problem is NP-hard, and we also present an approximation algorithm for the problem, in the case when k is a fixed constant. The approximation algorithm runs in time O(n log n) and produces a partition that is within a factor (4/3+ε) of the optimal if k=2, and a factor (2+ε) otherwise.


2011 ◽  
Vol 403-408 ◽  
pp. 2577-2580
Author(s):  
Wei Hong Xu ◽  
Min Zhu ◽  
Ya Ruo Jiang ◽  
Yu Shan Bai ◽  
Yan Yu

In this paper we present a spectral clustering method based on the MSRD (Most Similar Relation Diagram). The feature of this method is that both the constructing of the adjacency matrix and the clustering are achieved by spectral algorithm. Experiment on an artificial datasets demonstrate that our method can generate balanced partition and detect the manifold clusters no matter the unnormalized or normalized Laplassian is used and can generate partitions with different features if different MSRD is used. Experiments on some real datasets proved that our method is valid and effective.


2021 ◽  
Author(s):  
Xianyue Li ◽  
Yufei Pang ◽  
Chenxia Zhao ◽  
Yang Liu ◽  
Qingzhen Dong

Abstract Graph partition is a classical combinatorial optimization and graph theory problem, and it has a lot of applications, such as scientific computing, VLSI design and clustering etc. In this paper, we study the partition problem on large scale directed graphs under a new objective function, a new case of graph partition problem. We firstly propose the modeling of this problem, then design an algorithm based on multi-level strategy and recursive partition method, and finally do a lot of simulation experiments. The experimental results verify the stability of our algorithm and show that our algorithm has the same good performance as METIS. In addition, our algorithm is better than METIS on unbalanced ratio.


Author(s):  
Diego Recalde ◽  
Daniel Severín ◽  
Ramiro Torres ◽  
Polo Vaca

Sign in / Sign up

Export Citation Format

Share Document