Filters








57 Hits in 10.3 sec

GRAPH/Z: A Key-Value Store Based Scalable Graph Processing System

Tonglin Li, Chaoqi Ma, Jiabao Li, Xiaobing Zhou, Ke Wang, Dongfang Zhao, Iman Sadooghi, Ioan Raicu
2015 2015 IEEE International Conference on Cluster Computing  
Our system is built on top of ZHT, a scalable distributed key-value store, which benefits the graph processing in terms of scalability, performance and persistency.  ...  This work describes the design and implementation of a new graph processing system based on Bulk Synchronous Parallel model.  ...  Graph/Z can be considered as another Pregel-liked graph processing system, but it inherits some important features from ZHT, a distributed key-value store, which differentiate Graph/Z from other systems  ... 
doi:10.1109/cluster.2015.90 dblp:conf/cluster/LiMLZWZSR15 fatcat:dmhqihc2y5erpffofqsvnbu5jy

Analyzing temporal metrics of public transportation for designing scalable delay-tolerant networks

Adriano Galati, Vladimir Vukadinovic, Maria Olivares, Stefan Mangold
2013 Proceedings of the 8th ACM workshop on Performance monitoring and measurement of heterogeneous wireless and wired networks - PM2HW2N '13  
Temporal metrics of a public transportation network are determined, together with the overall value of exploiting the mobility of participating vehicles.  ...  Our analysis provides guidelines to design and build a hierarchical topological structure of similar networks.  ...  We have processed the complete temporal graph Z w t during peak and off-peak hours with directed links, using windows of five minutes and h=1.  ... 
doi:10.1145/2512840.2512846 dblp:conf/mswim/GalatiVOM13 fatcat:opl5ohhsgjg7dgxiozzgh75f54

The Complexity of Andersen's Analysis in Practice [chapter]

Manu Sridharan, Stephen J. Fink
2009 Lecture Notes in Computer Science  
variable and a sparse flow graph.  ...  In particular, we show that a simple algorithm can compute Andersen's analysis in worst-case quadratic time as long as the input program is k-sparse, i.e., it has at most k statements dereferencing each  ...  Andersen's analysis can be formulated as a dynamic transitive closure problem over a flow graph representing the flow of pointers in the program, where the flow graph grows dynamically to capture value  ... 
doi:10.1007/978-3-642-03237-0_15 fatcat:drvgvornknhlvabzp5eqldoenu

DEX: Self-healing Expanders [article]

Gopal Pandurangan, Peter Robinson, Amitabh Trehan
2013 arXiv   pre-print
We present a fully-distributed self-healing algorithm DEX, that maintains a constant degree expander network in a dynamic setting.  ...  The algorithm requires only a constant number of topology changes.  ...  We first look at the case where no staggered inflation/deflation is in progress: If some node u wants to store a key value pair (k, val) in the DHT, u computes the index z := h s (k) .  ... 
arXiv:1206.1522v4 fatcat:7mocj43xwfgkrlg5i7m6uilvwi

Local search in unstructured networks [chapter]

Lada A. Adamic, Rajan M. Lukose, Bernardo A. Huberman
2004 Handbook of Graphs and Networks  
Each node stores a routing table of files identified by a unique key and the node which is storing the file.  ...  The method is therefore parameterized by a sequence of ttl values and a waiting value.  ... 
doi:10.1002/3527602755.ch13 fatcat:fannre7tqzdv3nj2zmxwibsotu

Local Search in Unstructured Networks [article]

Lada A. Adamic, Rajan M. Lukose, Bernardo A. Huberman
2002 arXiv   pre-print
We review a number of message-passing algorithms that can be used to search through power-law networks.  ...  Most of these algorithms are meant to be improvements for peer-to-peer file sharing systems, and some may also shed some light on how unstructured social networks with certain topologies might function  ...  Each node stores a routing table of files identified by a unique key and the node which is storing the file.  ... 
arXiv:cond-mat/0204181v2 fatcat:ymdqcsynvbgxrjjdpgvero4iei

DEX: Self-Healing Expanders

Gopal Pandurangan, Peter Robinson, Amitabh Trehan
2014 2014 IEEE 28th International Parallel and Distributed Processing Symposium  
We first look at the case where no staggered inflation/deflation is in progress: If some node u wants to store a key value pair (k, val) in the DHT, u computes the index z := h s (k).  ...  Recall that, after Phase 1 all nodes have computed the same graph Z t ( p i+1 ) and thus use the same value σ .  ... 
doi:10.1109/ipdps.2014.78 dblp:conf/ipps/PanduranganRT14 fatcat:tmtmnazbuvaghmt7ozvnmgxw4i

Novel Approaches for Exclusive and Continuous Fingerprint Classification [chapter]

Javier A. Montoya-Zegarra, João P. Papa, Neucimar J. Leite, Ricardo da Silva Torres, Alexandre X. Falcão
2009 Lecture Notes in Computer Science  
measures used for fingerprint matching purposes, and (2) a novel multi-class object recognition method based on the Optimum Path Forest classifier.  ...  Experiments were carried out on the standard NIST-4 dataset aiming to study the discriminative and scalability capabilities of our approaches.  ...  By computing an MST in the complete graph (Z 1 , A) , we obtain a connected acyclic graph whose nodes are all samples in Z 1 and the arcs are undirected and weighted by the distance d between the adjacent  ... 
doi:10.1007/978-3-540-92957-4_34 fatcat:26klh57wrjh73m6wr7hwyursfy

A Multiple Pheromone Ant Clustering Algorithm [chapter]

Jan Chircop, Christopher D. Buckingham
2014 Studies in Computational Intelligence  
The value of the best function result so far is stored in a variable, p best n , which is then compared with later iterations.  ...  The next algorithm introduces the more structured graph-theoretic clustering. Graph-theoretic Clustering approaches In graph-based clustering, each data point is taken to be a node of a graph.  ... 
doi:10.1007/978-3-319-01692-4_2 fatcat:mvkv4ssuzbhevdqvyx3lgcnnxu

Rethinking and Scaling Up Graph Contrastive Learning: An Extremely Efficient Approach with Group Discrimination [article]

Yizhen Zheng, Shirui Pan, Vincent Cs Lee, Yu Zheng, Philip S. Yu
2022 arXiv   pre-print
paradigm for self-supervised GRL, namely, Group Discrimination (GD), and propose a novel GD-based method called Graph Group Discrimination (GGD).  ...  Graph contrastive learning (GCL) alleviates the heavy reliance on label information for graph representation learning (GRL) via self-supervised learning schemes.  ...  However, the diffusion augmentation involves expensive matrix inversion computation and significantly densifies the given graph, which requires much more memory and time to store and process than the original  ... 
arXiv:2206.01535v1 fatcat:5w3p5um4frgrvcjmom7fkaxa4q

Neural Subgraph Isomorphism Counting [article]

Xin Liu, Haojie Pan, Mutian He, Yangqiu Song, Xin Jiang, Lifeng Shang
2020 arXiv   pre-print
To make it scalable for large-scale graphs and patterns, we propose a learning framework which augments different representation learning architectures and iteratively attends pattern and target data graphs  ...  In this paper, we study a new graph learning problem: learning to count subgraph isomorphisms.  ...  We consider to use memory for the key and the value. Two gated multi-head attention is applied onǦ anď P, which is described in Appendix D.2.  ... 
arXiv:1912.11589v3 fatcat:crarue5fxbflfbmbma7mw3offa

Search in power-law networks

Lada A. Adamic, Rajan M. Lukose, Amit R. Puniyani, Bernardo A. Huberman
2001 Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics  
We introduce a number of local search strategies which utilize high degree nodes in power-law graphs and which have costs which scale sub-linearly with the size of the graph.  ...  Many communication and social networks have power-law link distributions, containing a few nodes which have a very high degree and many with low degree.  ...  Gnutella Gnutella is a peer-to-peer filesharing system which treats all client nodes as functionally equivalent and lacks a central server which can store file location information.  ... 
doi:10.1103/physreve.64.046135 pmid:11690118 fatcat:4n7ryueds5dpth6uc7zjk2h4je

Search in power-law networks [chapter]

L. A. Adamic, R. M. Lukose, A. R. Puniyani, B. A. Huberman
2011 The Structure and Dynamics of Networks  
We introduce a number of local search strategies which utilize high degree nodes in power-law graphs and which have costs which scale sub-linearly with the size of the graph.  ...  Many communication and social networks have power-law link distributions, containing a few nodes which have a very high degree and many with low degree.  ...  Gnutella Gnutella is a peer-to-peer filesharing system which treats all client nodes as functionally equivalent and lacks a central server which can store file location information.  ... 
doi:10.1515/9781400841356.543 fatcat:4uq4ay2fnbb2hj572yisvy35ii

Timed Hazard Analysis of Self-healing Systems [chapter]

Claudia Priesterjahn, Dominik Steenken, Matthias Tichy
2013 Lecture Notes in Computer Science  
Further, we demonstrate the scalability of the approach by experiments. 4 A Timed Failure Propagation Graph specifies the propagation of failures through components and the propagation times of failures  ...  These systems may react to failures by a structural reconfiguration of the architecture during runtime.  ...  We define a configuration as a typed graph as defined in [11] based on the type graph of the component specification.  ... 
doi:10.1007/978-3-642-36249-1_5 fatcat:2g6zm4cvc5fqzfpdpsj6j4kkvu

A survey of repositories in graph theory [article]

Srinibas Swain, C. Paul Bonnington, Graham Farr, Kerri Morgan
2020 arXiv   pre-print
Foster in the 1930s, many graph repositories have been created to support research in graph theory.  ...  This survey reviews many of these graph repositories and summarises the scope and contents of each repository.  ...  A graph parameter is defined as a function f : {graphs} → Z that is invariant under isomorphism. We sometimes extend the term to include non-integer-values like complex numbers and polynomials.  ... 
arXiv:2006.12741v1 fatcat:uxzb3xfahbesbdhgynn3y7kugu
« Previous Showing results 1 — 15 out of 57 results