Filters








99 Hits in 2.8 sec

How Well Do Graph-Processing Platforms Perform? An Empirical Performance Evaluation and Analysis

Yong Guo, Marcin Biczak, Ana Lucia Varbanescu, Alexandru Iosup, Claudio Martella, Theodore L. Willke
2014 2014 IEEE 28th International Parallel and Distributed Processing Symposium  
To alleviate this challenge, we propose an empirical method for benchmarking graph-processing platforms.  ...  Thus, users face the daunting challenge of selecting an appropriate platform for their specific application.  ...  GraphLab stores the entire graph and all program state in memory.  ... 
doi:10.1109/ipdps.2014.49 dblp:conf/ipps/GuoBVIMW14 fatcat:rqes3axydvdl7cejaek72dzjem

Pregelix: Big(ger) Graph Analytics on A Dataflow Engine [article]

Yingyi Bu, Vinayak Borkar, Jianfeng Jia, Michael J. Carey, Tyson Condie
2014 arXiv   pre-print
There is a growing need for distributed graph processing systems that are capable of gracefully scaling to very large graph datasets.  ...  ., we have seen up to 15x speedup compared to Apache Giraph and up to 35x speedup compared to distributed GraphLab), and makes more effective use of available machine resources to support Big(ger) Graph  ...  Finally, we thank Raghu Ramakrishnan for the early discussion of this work as well as the sponsorship for our access to the Yahoo! cluster for scale-testing early versions of the system.  ... 
arXiv:1407.0455v1 fatcat:gq6jqcmptbejjfslnee5wdaouu

Pregelix

Yingyi Bu, Vinayak Borkar, Jianfeng Jia, Michael J. Carey, Tyson Condie
2014 Proceedings of the VLDB Endowment  
There is a growing need for distributed graph processing systems that are capable of gracefully scaling to very large graph datasets.  ...  ., we have seen up to 15× speedup compared to Apache Giraph and up to 35× speedup compared to distributed GraphLab), and more effective use of available machine resources to support Big(ger) Graph Analytics  ...  Finally, we thank Raghu Ramakrishnan for discussing this work with us and sponsoring our access to a Yahoo! cluster for scale-testing early versions of the system.  ... 
doi:10.14778/2735471.2735477 fatcat:bx5bfbo24fhdnnxnpvyfnkn2ci

LFGraph

Imranul Hoque, Indranil Gupta
2013 Proceedings of the First ACM SIGOPS Conference on Timely Results in Operating Systems - TRIOS '13  
We present LFGraph, a fast, scalable, distributed, in-memory graph analytics engine intended primarily for directed graphs. LFGraph is the first system to satisfy all of the above requirements.  ...  Distributed graph analytics frameworks must offer low and balanced communication and computation, low preprocessing overhead, low memory footprint, and scalability.  ...  Acknowledgements: We thank our shepherd Benjamin Wester and anonymous reviewers for their insightful comments and suggestions.  ... 
doi:10.1145/2524211.2524218 dblp:conf/sosp/HoqueG13 fatcat:nrls3wmvkbhglddqzsyetom44u

From "think like a vertex" to "think like a graph"

Yuanyuan Tian, Andrey Balmin, Severin Andreas Corsten, Shirish Tatikonda, John McPherson
2013 Proceedings of the VLDB Endowment  
This vertex-centric model is easy to program and has been proved useful for many graph algorithms.  ...  To meet the challenge of processing rapidly growing graph and network data created by modern applications, a number of distributed graph processing systems have emerged, such as Pregel and GraphLab.  ...  Giraph++ uses Netty or Hadoop RPC for communication, whereas GraphLab uses MPI) and the use of different programming languages (Giraph++ in Java and GraphLab in C++) contribute a great deal to the difference  ... 
doi:10.14778/2732232.2732238 fatcat:6xnz5bqxcvcu7o3rpbiw5qfq4i

BigSparse: High-performance external graph analytics [article]

Sang-Woo Jun, Andy Wright, Sizhuo Zhang, Shuotao Xu, Arvind
2017 arXiv   pre-print
for terabyte-size graphs with billions of vertices.  ...  In our experiments on a server with 32GB to 64GB of DRAM, BigSparse outperforms other in-memory and semi-external graph analytics systems for algorithms such as PageRank, BreadthFirst Search, and Betweenness-Centrality  ...  Many prominent graph analytics platforms, including Pregel [13] and GraphLab [11] , expose a vertex-centric programming model because of its ease of distributed execution.  ... 
arXiv:1710.07736v1 fatcat:3rlvy45dtzazjecsqiilg2vgeu

Beyond Triangles: A Distributed Framework for Estimating 3-profiles of Large Graphs [article]

Ethan R. Elenberg, Karthikeyan Shanmugam, Michael Borokhovich, Alexandros G. Dimakis
2015 arXiv   pre-print
Our algorithm is distributed and operates as a vertex program over the GraphLab PowerGraph framework.  ...  Our algorithm uses the novel concept of 3-profile sparsifiers: sparse graphs that can be used to approximate the full 3-profile counts for a given large graph.  ...  Pregel, GraphLab, Galois, GraphX, see [30] for a comparison) are frameworks for expressing distributed computation on graphs in the language of vertex programs.  ... 
arXiv:1506.06671v1 fatcat:hr4pbfp6ezffjiwz7sfkz6xmzq

Beyond Triangles

Ethan R. Elenberg, Karthikeyan Shanmugam, Michael Borokhovich, Alexandros G. Dimakis
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
Our algorithm is distributed and operates as a vertex program over the GraphLab framework.  ...  Our algorithm uses the novel concept of 3-profile sparsifiers: sparse graphs that can be used to approximate the full 3-profile counts for a given large graph.  ...  Pregel, GraphLab, Galois, GraphX, see [30] for a comparison) are frameworks for expressing distributed computation on graphs in the language of vertex programs.  ... 
doi:10.1145/2783258.2783413 dblp:conf/kdd/ElenbergSBD15 fatcat:ccyhwydbebe5jbifh3juuo5fjq

Distributed Programming over Time-Series Graphs

Yogesh Simmhan, Neel Choudhury, Charith Wickramaarachchi, Alok Kumbhare, Marc Frincu, Cauligi Raghavendra, Viktor Prasanna
2015 2015 IEEE International Parallel and Distributed Processing Symposium  
We examine storage optimizations for GoFS, design patterns in Gopher to leverage the distributed data layout, and evaluate the GoFFish platform using time-series graph data and applications on a commodity  ...  Gopher is co-designed with GoFS, a distributed storage specialized for time-series graphs, as part of the GoFFish distributed analytics platform.  ...  Google recently proposed the Pregel model [8] of vertex centric programming for large scale graph processing, which also has similarities with GraphLab [18] .  ... 
doi:10.1109/ipdps.2015.66 dblp:conf/ipps/SimmhanCWKFRP15 fatcat:3kvc46tdbjftjohwyxmy2cptea

Big Graph Analytics Platforms

Da Yan, Yingyi Bu, Yuanyuan Tian, Amol Deshpande
2017 Foundations and Trends in Databases  
First, there is often interest in doing temporal analysis over historical traces of graphs, often called time-evolving graphs or historical graphs; examples of such analysis tasks include network evolution  ...  GraphChi GraphChi was proposed as a single-PC counterpart to distributed GraphLab, which keeps the GAS programming model but eliminates the requirement of a cluster of machines with large cumulative main  ...  Two thread pools are maintained in TurboGraph, one for the execution threads, and the other for the asynchronous I/O callback threads.  ... 
doi:10.1561/1900000056 fatcat:ucqrtzo4q5g2lpj6dmp7jv3e5m

Intel "big data" science and technology center vision and execution plan

Michael Stonebraker, Sam Madden, Pradeep Dubey
2013 SIGMOD record  
This paper presents the big data vision of this technology center and the execution plan for the first few years.  ...  Intel held a national competition for a 5th Science and Technology center in 2012 and selected a proposal from M.I.T. with a theme of "Big Data".  ...  We are also exploring visualization of large data sets and efficient algorithms for complex analysis.  ... 
doi:10.1145/2481528.2481537 fatcat:cfufdakydbfkfdjk2yftm572la

Characterizing Application Memory Error Vulnerability to Optimize Datacenter Cost via Heterogeneous-Reliability Memory

Yixin Luo, Sriram Govindan, Bikash Sharma, Mark Santaniello, Justin Meza, Aman Kansal, Jie Liu, Badriddine Khessib, Kushagra Vaid, Onur Mutlu
2014 2014 44th Annual IEEE/IFIP International Conference on Dependable Systems and Networks  
Existing approaches to providing reliability for memory devices pessimistically treat all data as equally vulnerable to memory errors.  ...  This presents an opportunity to greatly reduce server hardware cost by provisioning the right amount of memory reliability for different applications.  ...  We thank the anonymous reviewers and the members of SAFARI research group for feedback. We acknowledge the support of Microsoft and Samsung.  ... 
doi:10.1109/dsn.2014.50 dblp:conf/dsn/LuoGSSMKLKVM14 fatcat:57c4bkj3brha7jbzxosrrfwucu

Toward Scalable Systems for Big Data Analytics: A Technology Tutorial

Han Hu, Yonggang Wen, Tat-Seng Chua, Xuelong Li
2014 IEEE Access  
More complex analysis for production workload traces can be found in the authors' subsequent research [289] . Ghazal et al.  ...  GraphLab proposes three consistency models, full, edge, and vertex consistency, to allow for different levels of parallelism. • Stream Processing Model: S4 [36] and Storm [35] are two distributed stream  ... 
doi:10.1109/access.2014.2332453 fatcat:6unxlocbmnhs7mfctlvphdyqtu

Systems and Algorithms for Large-scale Graph Analytics (Dagstuhl Seminar 14462)

Eiko Yoneki, Amitabha Roy, Derek Murray, Marc Herbstritt
2015 Dagstuhl Reports  
This report documents the program and the outcomes of Dagstuhl Seminar 14462 "Systems and Algorithms for Large-scale Graph Analytics".  ...  We also thank Karthik Nilakant and Valentin Dalibard for their help with editing the report.  ...  We thank Derek Murray for his great contribution to the organization of the workshop and participation via Skype as he could not travel because of the sudden set down of Microsoft Research Silicon Valley  ... 
doi:10.4230/dagrep.4.11.59 dblp:journals/dagstuhl-reports/Yoneki0M14 fatcat:dilicps65jgipetfid3udj2q4i

Views and Transactional Storage for Large Graphs [chapter]

Michael M. Lee, Indrajit Roy, Alvin AuYoung, Vanish Talwar, K. R. Jayaram, Yuanyuan Zhou
2013 Lecture Notes in Computer Science  
We present Concerto, a graph store based on distributed, in-memory data structures.  ...  Using graph views, programmers can perform event-driven analysis and dynamically optimize application performance.  ...  We thank the anonymous reviewers for their valuable feedback. Part of this research was sponsored by the DARPA GRAPHS program (BAA-12-01).  ... 
doi:10.1007/978-3-642-45065-5_15 fatcat:vvzx3ptnnje5xivuzpmwgfxyze
« Previous Showing results 1 — 15 out of 99 results