#### DMCA

## Incremental Closeness Centrality in Distributed Memory (2015)

### Citations

532 | A faster algorithm for betweenness centrality
- Brandes
- 2001
(Show Context)
Citation Context .... As the networks become large, efficiency becomes a crucial concern while analyzing these networks. The algorithm with the best asymptotic complexity to compute the closeness and betweenness metrics =-=[4]-=- is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms [7, 9, 24] and high performance computing techniques [22, 32, 20]. ... |

204 | Mapping networks of terrorist cells
- Krebs
(Show Context)
Citation Context ...ions. They have been successfully used to carry analysis for various purposes such as power grid contingency analysis [16], quantifying importance in social networks [23], analysis of covert networks =-=[18]-=-, decision/action networks [8], and even for finding the best store locations in cities [26]. As the networks become large, efficiency becomes a crucial concern while analyzing these networks. The alg... |

150 | OSKI: A library of automatically tuned sparse matrix kernels
- Vuduc, Demmel, et al.
- 2005
(Show Context)
Citation Context ...tly hindered by irregular memory accesses. The most famous example for sparse computation is the multiplication of a sparse matrix by a dense vector (SpMV). Several techniques, like register blocking =-=[6, 33]-=- and usage of different matrix storage formats [2, 21], are proposed to regularize the memory access pattern. However, multiplying a sparse matrix by multiple vectors is the most efficient and popular... |

86 |
Algorithm 447: Efficient algorithms for graph manipulation
- Hopcroft, Tarjan
- 1973
(Show Context)
Citation Context ... If the intersection is empty (see the addition of bd in Figure 3(b)), we construct Π′ from scratch and set cid = Π′(uv) (e.g., cid = 2 in Figure 3(c)). A BCD can be computed in linear, O(m + n) time =-=[14]-=-. Hence, the cost of BCD maintenance is negligible compared to the cost of updating closeness centrality. Let G′cid = (Vcid, E ′ cid) be the biconnected component of G ′ containing uv. Let Acid ⊆ Vcid... |

70 |
Distributed processing of very large datasets with datacutter. Parallel Computing 27(11):1457
- Beynon, Kurc, et al.
- 2001
(Show Context)
Citation Context .... The paper is organized as follows: Section 2 introduces the notation, formally defines the closeness centrality metric, and describes the incremental approach in [27]. Section 3 presents DataCutter =-=[3]-=-, our in-house distributed memory dataflow middleware leveraged in 4 this work. Section 4 describes the proposed distributed framework for incremental centrality computations in detail. The experiment... |

54 | Fast approximation of centrality
- Eppstein, Wang
(Show Context)
Citation Context ...tic complexity to compute the closeness and betweenness metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms =-=[7, 9, 24]-=- and high performance computing techniques [22, 32, 20]. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-b... |

52 | The combinatorial blas: design, implementation, and applications
- Buluç, Gilbert
(Show Context)
Citation Context ... of any level `′ < `, then vertex v is part of level `. This formulation is used in parallel implementations of BFS on GPU [15, 25, 32], on shared memory systems [1] and on distributed memory systems =-=[5]-=-. 10 The algorithm is better represented using binary variables. Let x`i be the binary variable that is true if vertex i is part of the frontier at level ` for a BFS. The neighbors of level ` is repre... |

49 | A faster parallel algorithm and efficient multithreaded implementations for evaluating betweenness centrality on massive datasets
- Madduri, Ediger, et al.
(Show Context)
Citation Context ...ss metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms [7, 9, 24] and high performance computing techniques =-=[22, 32, 20]-=-. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-based analysis. Many of today’s networks are dynamic. An... |

40 | Scalable graph exploration on multicore processors
- Agarwal, Petrini, et al.
- 2010
(Show Context)
Citation Context ...s part of level `− 1 and v is not part of any level `′ < `, then vertex v is part of level `. This formulation is used in parallel implementations of BFS on GPU [15, 25, 32], on shared memory systems =-=[1]-=- and on distributed memory systems [5]. 10 The algorithm is better represented using binary variables. Let x`i be the binary variable that is true if vertex i is part of the frontier at level ` for a ... |

38 | Biomedical Image Analysis on a Cooperative Cluster of GPUs and Multicores
- Hartley, Catalyurek, et al.
(Show Context)
Citation Context ...ave different responsibilities and operate on data passing through them. DataCutter follows the component-based programming paradigm which has been used to describe and implement complex applications =-=[11, 12, 13, 29]-=- by way of components - distinct tasks with well-defined interfaces. This is also known as the filter-stream programming model [3] (a specific implementation of the dataflow programming model). A stre... |

30 | Skill characterization based on betweenness
- Şimşek, Barto
- 2009
(Show Context)
Citation Context ...ly used to carry analysis for various purposes such as power grid contingency analysis [16], quantifying importance in social networks [23], analysis of covert networks [18], decision/action networks =-=[8]-=-, and even for finding the best store locations in cities [26]. As the networks become large, efficiency becomes a crucial concern while analyzing these networks. The algorithm with the best asymptoti... |

24 | Ranking of closeness centrality for large-scale social networks
- Okamoto, Chen, et al.
- 2008
(Show Context)
Citation Context ...tic complexity to compute the closeness and betweenness metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms =-=[7, 9, 24]-=- and high performance computing techniques [22, 32, 20]. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-b... |

21 | Reduced-bandwidth multithreaded algorithms for sparse matrix-vector multiplication
- Buluç, Williams, et al.
(Show Context)
Citation Context ...tly hindered by irregular memory accesses. The most famous example for sparse computation is the multiplication of a sparse matrix by a dense vector (SpMV). Several techniques, like register blocking =-=[6, 33]-=- and usage of different matrix storage formats [2, 21], are proposed to regularize the memory access pattern. However, multiplying a sparse matrix by multiple vectors is the most efficient and popular... |

20 |
A novel application of parallel betweenness centrality to power grid contingency analysis
- Jin, Huang, et al.
- 2010
(Show Context)
Citation Context ...ng an entity dissemination? Centrality metrics have been used to answer such questions. They have been successfully used to carry analysis for various purposes such as power grid contingency analysis =-=[16]-=-, quantifying importance in social networks [23], analysis of covert networks [18], decision/action networks [8], and even for finding the best store locations in cities [26]. As the networks become l... |

18 |
A fast algorithm for streaming betweenness centrality
- Green, McColl, et al.
- 2012
(Show Context)
Citation Context ...rform centrality-based analysis. Many of today’s networks are dynamic. And for such networks, maintaining the exact centrality scores is a challenging problem which has been studied in the literature =-=[10, 19, 27]-=-. The problem can also arise for applications involving static networks such as the power grid contingency analysis and robustness evaluation of a network. The findings of such analyses and evaluation... |

18 | A quick algorithm for updating betweenness centrality
- Lee, Lee, et al.
- 2012
(Show Context)
Citation Context ...rform centrality-based analysis. Many of today’s networks are dynamic. And for such networks, maintaining the exact centrality scores is a challenging problem which has been studied in the literature =-=[10, 19, 27]-=-. The problem can also arise for applications involving static networks such as the power grid contingency analysis and robustness evaluation of a network. The findings of such analyses and evaluation... |

15 | Betweenness centrality: algorithms and lower bounds. arXiv
- Kintali
- 2008
(Show Context)
Citation Context ...comes a crucial concern while analyzing these networks. The algorithm with the best asymptotic complexity to compute the closeness and betweenness metrics [4] is believed to be asymptotically optimal =-=[17]-=-. The research on fast centrality computation have focused on approximation algorithms [7, 9, 24] and high performance computing techniques [22, 32, 20]. Today, the networks to be analyzed can be quit... |

15 |
Street centrality and densities of retail and services
- Porta, Latora, et al.
- 2009
(Show Context)
Citation Context ...grid contingency analysis [16], quantifying importance in social networks [23], analysis of covert networks [18], decision/action networks [8], and even for finding the best store locations in cities =-=[26]-=-. As the networks become large, efficiency becomes a crucial concern while analyzing these networks. The algorithm with the best asymptotic complexity to compute the closeness and betweenness metrics ... |

15 |
Fast network centrality analysis using GPUs
- Shi, Zhang
- 2011
(Show Context)
Citation Context ...ss metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms [7, 9, 24] and high performance computing techniques =-=[22, 32, 20]-=-. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-based analysis. Many of today’s networks are dynamic. An... |

14 |
Pattern-based sparse matrix representation for memory-efficient SMVM kernels
- Belgin, Back, et al.
- 2009
(Show Context)
Citation Context ...mous example for sparse computation is the multiplication of a sparse matrix by a dense vector (SpMV). Several techniques, like register blocking [6, 33] and usage of different matrix storage formats =-=[2, 21]-=-, are proposed to regularize the memory access pattern. However, multiplying a sparse matrix by multiple vectors is the most efficient and popular technique to regularize the memory access pattern. On... |

12 | Efficient sparse matrix-vector multiplication on x86-based many-core processors
- Liu, Smelyanskiy, et al.
(Show Context)
Citation Context ...mous example for sparse computation is the multiplication of a sparse matrix by a dense vector (SpMV). Several techniques, like register blocking [6, 33] and usage of different matrix storage formats =-=[2, 21]-=-, are proposed to regularize the memory access pattern. However, multiplying a sparse matrix by multiple vectors is the most efficient and popular technique to regularize the memory access pattern. On... |

11 |
Fast centrality approximation in modular networks
- Chan, Leung, et al.
(Show Context)
Citation Context ...tic complexity to compute the closeness and betweenness metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms =-=[7, 9, 24]-=- and high performance computing techniques [22, 32, 20]. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-b... |

11 | Centralities: Capturing the fuzzy notion of importance in social graphs
- Merrer, Trédan
- 2009
(Show Context)
Citation Context ...ave been used to answer such questions. They have been successfully used to carry analysis for various purposes such as power grid contingency analysis [16], quantifying importance in social networks =-=[23]-=-, analysis of covert networks [18], decision/action networks [8], and even for finding the best store locations in cities [26]. As the networks become large, efficiency becomes a crucial concern while... |

9 | DisNet: A framework for distributed graph computation
- Lichtenwalter, Chawla
- 2011
(Show Context)
Citation Context ...ss metrics [4] is believed to be asymptotically optimal [17]. The research on fast centrality computation have focused on approximation algorithms [7, 9, 24] and high performance computing techniques =-=[22, 32, 20]-=-. Today, the networks to be analyzed can be quite large, and we are always in a quest for faster techniques which help us to perform centrality-based analysis. Many of today’s networks are dynamic. An... |

7 | Investigating the use of gpu-accelerated nodes for sar image formation
- Hartley, Fasih, et al.
- 2009
(Show Context)
Citation Context ...ave different responsibilities and operate on data passing through them. DataCutter follows the component-based programming paradigm which has been used to describe and implement complex applications =-=[11, 12, 13, 29]-=- by way of components - distinct tasks with well-defined interfaces. This is also known as the filter-stream programming model [3] (a specific implementation of the dataflow programming model). A stre... |

6 |
Edge vs. node parallelism for graph centrality metrics
- Jia, Lu, et al.
- 2011
(Show Context)
Citation Context ...nous BFS: If one of the neighbor of v is part of level `− 1 and v is not part of any level `′ < `, then vertex v is part of level `. This formulation is used in parallel implementations of BFS on GPU =-=[15, 25, 32]-=-, on shared memory systems [1] and on distributed memory systems [5]. 10 The algorithm is better represented using binary variables. Let x`i be the binary variable that is true if vertex i is part of ... |

6 |
Computing betweenness centrality for small world networks on a GPU,” Poster at 15th High Performance Embedded Computing Conference
- Pande, Bader
- 2011
(Show Context)
Citation Context ...nous BFS: If one of the neighbor of v is part of level `− 1 and v is not part of any level `′ < `, then vertex v is part of level `. This formulation is used in parallel implementations of BFS on GPU =-=[15, 25, 32]-=-, on shared memory systems [1] and on distributed memory systems [5]. 10 The algorithm is better represented using binary variables. Let x`i be the binary variable that is true if vertex i is part of ... |

5 |
Regularizing graph centrality computations
- Sarıyüce, Saule, et al.
(Show Context)
Citation Context ...g processed. Thus, the traditional queue-based implementation of BFS does not seem to be easily extendable to support concurrent BFSs (co-BFS) in a vector-friendly manner. We developed this method in =-=[30, 31]-=- and present here the main idea. 2.5.1. An SpMV-based formulation of closeness centrality The idea is to convert to a simpler definition of level synchronous BFS: If one of the neighbor of v is part o... |

4 |
Çatalyürek, “Shattering and compressing networks for betweenness centrality
- Sariyüce, Saule, et al.
- 2013
(Show Context)
Citation Context ...rse asymptotic complexity the performance of co-BFS outperforms traditional BFS approach [30, 31]. Moreover, such algorithm is compatible with the decomposition of the graph in biconnected components =-=[28]-=- which can lead to further improvement. Because this algorithm computes the farness of the sources, it 11 can be used to compute centrality incrementally. 3. DataCutter Streamer employs DataCutter [3]... |

4 | STREAMER: a distributed framework for incremental closeness centrality computation
- Sarıyüce, Saule, et al.
- 2013
(Show Context)
Citation Context ...ces are colored with respect to their relative CC scores where red implies a higher closeness score. the algorithm was not fast enough to be used in practice. In a previous work, we proposed Streamer =-=[29]-=- to parallelize these incremental algorithms. In this paper, we present an improved version of Streamer, to efficiently parallelize the incremental CC computation on high-performance clusters. The bes... |

3 |
Improving performance of adaptive component-based dataflow middleware
- Hartley, Saule, et al.
- 2012
(Show Context)
Citation Context ...ave different responsibilities and operate on data passing through them. DataCutter follows the component-based programming paradigm which has been used to describe and implement complex applications =-=[11, 12, 13, 29]-=- by way of components - distinct tasks with well-defined interfaces. This is also known as the filter-stream programming model [3] (a specific implementation of the dataflow programming model). A stre... |

2 | Hardware/software vectorization for closeness centrality on multi-/many-core architectures
- Sarıyüce, Saule, et al.
- 2014
(Show Context)
Citation Context ...g processed. Thus, the traditional queue-based implementation of BFS does not seem to be easily extendable to support concurrent BFSs (co-BFS) in a vector-friendly manner. We developed this method in =-=[30, 31]-=- and present here the main idea. 2.5.1. An SpMV-based formulation of closeness centrality The idea is to convert to a simpler definition of level synchronous BFS: If one of the neighbor of v is part o... |

1 | Incremental algorithms for closeness centrality
- Sarıyüce, Kaya, et al.
- 2013
(Show Context)
Citation Context ...rform centrality-based analysis. Many of today’s networks are dynamic. And for such networks, maintaining the exact centrality scores is a challenging problem which has been studied in the literature =-=[10, 19, 27]-=-. The problem can also arise for applications involving static networks such as the power grid contingency analysis and robustness evaluation of a network. The findings of such analyses and evaluation... |