bibliography.bib

@inproceedings{spark-sql,
  title={Spark SQL: Relational data Processing in Spark},
  author={Armbrust, Michael and Xin, Reynold S and Lian, Cheng and Huai, Yin and Liu, Davies and Bradley, Joseph K and Meng, Xiangrui and Kaftan, Tomer and Franklin, Michael J and Ghodsi, Ali and others},
  booktitle={International Conference on Management of Data (SIGMOD)},
  pages={1383--1394},
  year={2015},
  organization={ACM}
}
@inproceedings{open-cypher,
  title={Cypher: An Evolving Query Language for Property Graphs},
  author={Francis, Nadime and Green, Alastair and Guagliardo, Paolo and Libkin, Leonid and Lindaaker, Tobias and Marsault, Victor and Plantikow, Stefan and Rydberg, Mats and Selmer, Petra and Taylor, Andr{\'e}s},
  booktitle={International Conference on Management of Data (SIGMOD)},
  pages={1433--1445},
  year={2018},
  organization={ACM}
}
@inproceedings{gcore,
  title={G-CORE: A Core for Future Graph Query Languages},
  author={Angles, Renzo and Arenas, Marcelo and Barcelo, Pablo and Boncz, Peter and Fletcher, George and Gutierrez, Claudio and Lindaaker, Tobias and Paradies, Marcus and Plantikow, Stefan and Sequeda, Juan and others},
  booktitle={International Conference on Management of Data (SIGMOD)},
  pages={1421--1432},
  year={2018},
  organization={ACM}
}
@inproceedings{olddog,
  title={Join Processing for Graph Patterns: An Old Dog with new Tricks},
  author={Nguyen, Dung and Aref, Molham and Bravenboer, Martin and Kollias, George and Ngo, Hung Q and R{\'e}, Christopher and Rudra, Atri},
  booktitle={GRADES},
  pages={2},
  year={2015},
  organization={ACM}
}
@article{lftj,
  title={Leapfrog Triejoin: A Simple, Worst-Case Optimal Join Algorithm},
  author={Veldhuizen, Todd L},
  journal={arXiv preprint arXiv:1210.0481},
  year={2012}
}
@inproceedings{myria-detailed,
  title={From Theory to Practice: Efficient Join Query Evaluation in a Parallel Database System},
  author={Chu, Shumo and Balazinska, Magdalena and Suciu, Dan},
  booktitle={International Conference on Management of Data (SIGMOD)},
  pages={63--78},
  year={2015},
  organization={ACM}
}
@mastersthesis{andreas,
  title={Evaluation of Worst-Case Optimal Join Algorithms},
  year={2017},
  school={Technische Universität München},
  author={Amler, Andreas}
}
@inproceedings{myria,
  title={Demonstration of the Myria Big Data Management Service},
  author={Halperin, Daniel and Teixeira de Almeida, Victor and Choo, Lee Lee and Chu, Shumo and Koutris, Paraschos and Moritz, Dominik and Ortiz, Jennifer and Ruamviboonsuk, Vaspol and Wang, Jingjing and Whitaker, Andrew and others},
  booktitle={International Conference on Management of Data (SIGMOD)},
  pages={881--884},
  year={2014},
  organization={ACM}
}
@inproceedings{graphx,
  title={GraphX: Graph Processing in a Distributed Dataflow Framework.},
  author={Gonzalez, Joseph E and Xin, Reynold S and Dave, Ankur and Crankshaw, Daniel and Franklin, Michael J and Stoica, Ion},
  booktitle={Symposium on Operating System Design and Implementation (ODSI)},
  volume={14},
  pages={599--613},
  year={2014},
  publisher = {Usenix},
}
@inproceedings{ammar2018distributed,
  title={Distributed Evaluation of Subgraph Queries Using Worst-Case Optimal Low-Memory Dataflows},
  author={Ammar, Khaled and McSherry, Frank and Salihoglu, Semih and Joglekar, Manas},
  booktitle={VLDB Endowment},
  volume={11},
  number={6},
  pages={691--704},
  year={2018},
}

@online{caps,
  author = {openCypher Project},
  title = {CAPS: Cypher for Apache Spark},
  year = 2016,
  url = {https://github.com/opencypher/cypher-for-apache-spark},
  urldate = {2019-02-18}
}

@online{tungsten-project,
  author = {Xin, Reynold and Rosen, Josh},
  title = {Project Tungsten: Bringing Apache Spark Closer to Bare Metal},
  year = 2015,
  url = {https://databricks.com/blog/2015/04/28/project-tungsten-bringing-spark-closer-to-bare-metal.html},
  urldate = {2019-09-26}
}

@online{tungsten-code-generation,
  author = {Agarwal, Sameer and Liu, Davies and Xin, Reynold},
  title = {Apache Spark as a Compiler: Joining a Billion Rows per Second on a Laptop, Deep dive into the new Tungsten execution engine},
  year = 2016,
  url = {https://databricks.com/blog/2016/05/23/apache-spark-as-a-compiler-joining-a-billion-rows-per-second-on-a-laptop.html},
  urldate = {2019-09-26}
}

@online{rdd-programming-guide,
  author = {Databricks},
  title = {Apache Spark Documentation: RDD Programming Guide},
  url = {https://spark.apache.org/docs/2.2.3/rdd-programming-guide.html},
  urldate = {2019-09-26}
}

@online{spark-internals-structured-query-execution,
  author = {Laskowski, Jacek},
  title = {The Internals of Spark SQL, QueryExecution - Structured Query Execution Pipeline},
  url = {https://jaceklaskowski.gitbooks.io/mastering-spark-sql/spark-sql-QueryExecution.html},
  urldate = {2019-09-26}
}


@inproceedings{ngo2012worst,
  title={Worst-Case Optimal Join Algorithms},
  author={Ngo, Hung Q and Porat, Ely and R{\'e}, Christopher and Rudra, Atri},
  booktitle={SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems},
  pages={37--48},
  year={2012},
  organization={ACM}
}
@inproceedings{naiad,
  title={Naiad: A Timely Dataflow System},
  author={Murray, Derek G and McSherry, Frank and Isaacs, Rebecca and Isard, Michael and Barham, Paul and Abadi, Mart{\'\i}n},
  booktitle={Symposium on Operating Systems Principles},
  pages={439--455},
  year={2013},
  organization={ACM}
}
@inproceedings{mapreduce,
  title	= {MapReduce: Simplified Data Processing on Large Clusters},
  author	= {Jeffrey Dean and Sanjay Ghemawat},
  year	= {2004},
  booktitle	= {Symposium on Operating System Design and Implementation (ODSI)},
  pages	= {137--150},
  address	= {San Francisco, CA},
  publisher = {Usenix},
}
@inproceedings{rdd,
  title={Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing},
  author={Zaharia, Matei and Chowdhury, Mosharaf and Das, Tathagata and Dave, Ankur and Ma, Justin and McCauley, Murphy and Franklin, Michael J and Shenker, Scott and Stoica, Ion},
  booktitle={Conference on Networked Systems Design and Implementation},
  pages={2--2},
  year={2012},
  organization={USENIX}
}

@inproceedings{yarn,
  title={Apache Hadoop Yarn: Yet Another Resource Negotiator},
  author={Vavilapalli, Vinod Kumar and Murthy, Arun C and Douglas, Chris and Agarwal, Sharad and Konar, Mahadev and Evans, Robert and Graves, Thomas and Lowe, Jason and Shah, Hitesh and Seth, Siddharth and others},
  booktitle={Symposium on Cloud Computing},
  pages={5},
  year={2013},
  organization={ACM}
}

@online{hadoop,
  url={http://hadoop.apache.org},
  title = {Apache hadoop},
  urldate = {2019-09-26}
}

@online{kubernetes,
  url={https://kubernetes.io},
  title = {kubernetes},
  urldate = {2019-09-26}
}

@inproceedings{mesos,
  title={Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center.},
  author={Hindman, Benjamin and Konwinski, Andy and Zaharia, Matei and Ghodsi, Ali and Joseph, Anthony D and Katz, Randy H and Shenker, Scott and Stoica, Ion},
  booktitle={NSDI},
  volume={11},
  number={2011},
  pages={22--22},
  year={2011}
}

@article{spark,
  title={Spark: Cluster Computing with Working Sets.},
  author={Zaharia, Matei and Chowdhury, Mosharaf and Franklin, Michael J and Shenker, Scott and Stoica, Ion},
  journal={HotCloud},
  volume={10},
  number={10-10},
  pages={95},
  year={2010}
}
@inproceedings{agm,
  title={Size Bounds and Query Plans for Relational Joins},
  author={Atserias, Albert and Grohe, Martin and Marx, D{\'a}niel},
  booktitle={Foundations of Computer Science (FOCS)},
  pages={739--748},
  year={2008},
  organization={IEEE}
}

@inproceedings{haloop,
  title={HaLoop: Efficient Iterative Data Processing on Large Clusters},
  author={Bu, Yingyi and Howe, Bill and Balazinska, Magdalena and Ernst, Michael D},
  journal={VLDB Endowment},
  volume={3},
  number={1-2},
  pages={285--296},
  year={2010},
  publisher={VLDB Endowment}
}

@inproceedings{twitter-diamond,
  title={Real-Time Twitter Recommendation: Online Motif Detection in Large Dynamic Graphs},
  author={Gupta, Pankaj and Satuluri, Venu and Grewal, Ajeet and Gurumurthy, Siva and Zhabiuk, Volodymyr and Li, Quannan and Lin, Jimmy},
  booktitle={VLDB Endowment},
  volume={7},
  number={13},
  pages={1379--1380},
  year={2014},
}
@inproceedings{nprr,
  title={Worst-Case Optimal Join Algorithms},
  author={Ngo, Hung Q and Porat, Ely and R{\'e}, Christopher and Rudra, Atri},
  booktitle={SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems},
  pages={37--48},
  year={2012},
  organization={ACM}
}
@online{neo4j,
  author = {Neo4j},
  title = {Neo4j},
  year = {2000},
  url = {https://neo4j.com/},
  urldate = {2019-02-18}
}
@online{janusgraph,
author = {},
title = {JanusGraph},
year = 2017,
url = {http://janusgraph.org/},
urldate = {2019-02-18}
}

@inproceedings{pregel,
title={Pregel: A System for Large-Scale Graph Processing},
author={Malewicz, Grzegorz and Austern, Matthew H and Bik, Aart JC and Dehnert, James C and Horn, Ilan and Leiser, Naty and Czajkowski, Grzegorz},
booktitle={International Conference on Management of Data (SIGMOD)},
pages={135--146},
year={2010},
organization={ACM}
}
@online{graphframe,
author = {Dave, Ankur and  Bradley, Joseph and Hunter, Tim and Meng, Xiangrui},
title = {GraphFrame},
year = 2016,
url = {https://databricks.com/blog/2016/03/03/introducing-graphframes.html},
urldate = {2019-02-18}
}
@online{myria-website,
author = {},
title = {Myria Project Website},
year = 2012,
url = {http://myria.cs.washington.edu/},
urldate = {2019-03-14}
}

@inproceedings{logicBlox,
title={Design and Implementation of the LogicBlox System},
author={Aref, Molham and ten Cate, Balder and Green, Todd J and Kimelfeld, Benny and Olteanu, Dan and Pasalic, Emir and Veldhuizen, Todd L and Washburn, Geoffrey},
booktitle={International Conference on Management of Data (SIGMOD)},
pages={1371--1382},
year={2015},
organization={ACM}
}

@inproceedings{graphflow,
title={Graphflow: An Active Graph Database},
author={Kankanamge, Chathura and Sahu, Siddhartha and Mhedbhi, Amine and Chen, Jeremy and Salihoglu, Semih},
booktitle={International Conference on Management of Data (SIGMOD)},
pages={1695--1698},
year={2017},
organization={ACM}
}

@article{emptyheaded,
title={Emptyheaded: A Relational Engine for Graph Processing},
author={Aberger, Christopher R and Lamb, Andrew and Tu, Susan and N{\"o}tzli, Andres and Olukotun, Kunle and R{\'e}, Christopher},
journal={Transactions on Database Systems (TODS)},
volume={42},
number={4},
pages={20},
year={2017},
publisher={ACM}
}

@online{leapfrog-triejoin-schroeder,
author = {Schroeder dewitt, Christian},
title = {Leapfrog Triejoin Implementation for `Database Systems and Implementation' at Oxford University},
year = 2012,
url = {https://github.com/schroeder-dewitt/leapfrog-triejoin},
urldate = {2019-03-14}
}

@online{snb,
author = {LDBC},
title = {LDBC SNB Documentation},
year = {2017},
url = {https://github.com/ldbc/ldbc_snb_docs},
urldate = {2019-04-10}
}

@online{databricks-scala-guide,
author = {Databricks},
title = {Databricks Scala Guide},
year = {2018},
url = {https://github.com/databricks/scala-style-guide/blob/7eb5477781c11f9a75a2d8d6ef773ca6965f4ea0/README.md},
urldate = {2019-05-25}
}
@article{skew-strikes-back,
title={Skew Strikes Back: New Developments in the Theory of Join Algorithms},
author={Ngo, Hung Q and R{\'e}, Christopher and Rudra, Atri},
journal={arXiv preprint arXiv:1310.3314},
year={2013}
}
@inproceedings{beame2013,
title={Communication Steps for Parallel Query Processing},
author={Beame, Paul and Koutris, Paraschos and Suciu, Dan},
booktitle={SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems},
pages={273--284},
year={2013},
organization={ACM}
}
@inproceedings{beame2014,
title={Skew in Parallel Query Processing},
author={Beame, Paul and Koutris, Paraschos and Suciu, Dan},
booktitle={SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems},
pages={212--223},
year={2014},
organization={ACM}
}
@inproceedings{beame2016,
title={Worst-Case Optimal Algorithms for Parallel Query Processing},
author={Koutris, Paraschos and Beame, Paul and Suciu, Dan},
booktitle={International Conference on Database Theory (ICDT)},
year={2016},
organization={Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik}
}

@misc{snapnets,
author       = {Jure Leskovec and Andrej Krevl},
title        = {{SNAP Datasets}: {Stanford} Large Network Dataset Collection},
howpublished = {\url{http://snap.stanford.edu/data}},
month        = jun,
year         = 2014
}



@article{shares,
title={Optimizing Multiway Joins in a Map-Reduce Environment},
author={Afrati, Foto N and Ullman, Jeffrey D},
journal={Transactions on Knowledge and Data Engineering (TKDE)},
volume={23},
number={9},
pages={1282--1298},
year={2011},
publisher={IEEE}
}

@article{shares-skew,
title={SharesSkew: An Algorithm to Handle Skew for Joins in MapReduce},
author={Afrati, Foto N and Stasinopoulos, Nikos and Ullman, Jeffrey D and Vassilakopoulos, Angelos},
journal={Information Systems},
volume={77},
pages={129--150},
year={2018},
publisher={Elsevier}
}

@inproceedings{shares-proof,
title={Skew in Parallel Query Processing},
author={Beame, Paul and Koutris, Paraschos and Suciu, Dan},
booktitle={SIGMOD-SIGACT-SIGART symposium on Principles of database systems},
pages={212--223},
year={2014},
organization={ACM}
}

@inproceedings{shares-binary,
title={A worst-case optimal multi-round algorithm for parallel computation of conjunctive queries},
author={Ketsman, Bas and Suciu, Dan},
booktitle={SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems},
pages={417--428},
year={2017},
organization={ACM}
}

@inproceedings{shares-skew-proof,
title={Worst-case Optimal Algorithms for Parallel Query Processing},
author={Koutris, Paraschos and Beame, Paul and Suciu, Dan},
booktitle={International Conference on Database Theory (ICDT)},
year={2016},
organization={Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik}
}

@inproceedings{gym,
title={GYM: A Multiround Distributed Join Algorithm},
author={Afrati, Foto N and Joglekar, Manas R and Re, Christopher M and Salihoglu, Semih and Ullman, Jeffrey D},
booktitle={20th International Conference on Database Theory (ICDT)},
year={2017},
organization={Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik}
}

@techreport{ppp-fuchs-2019,
title={Solving Rubik’s Cubes by Iterative Deepening Search},
author={Fuchs, Per},
year={2019},
institution={Vrije Universiteit Amsterdam}
}

@misc{lectureNotesPLPS,
author        = {Henri Bal},
title         = {Lecture Notes in Programming Large Scale Parallel Systems},
month         = {09},
year          = {2018},
publisher={Vrije Universiteit Amsterdam}
}



@inproceedings{cost,
title={Scalability! But at What {COST}?},
author={McSherry, Frank and Isard, Michael and Murray, Derek G},
booktitle={Hot Topics in Operating Systems (HotOS {XV})},
year={2015}
}

@article{snap,
title={{SNAP}: A General-Purpose Network Analysis and Graph-Mining Library},
author={Leskovec, Jure and Sosi{\v{c}}, Rok},
journal={Transactions on Intelligent Systems and Technology (TIST)},
volume={8},
number={1},
pages={1},
year={2016},
publisher={ACM}
}
@article{salihoglu2018,
title={Response to “Scale Up or Scale Out for Graph Processing”},
author={Salihoglu, Semih and {\"O}zsu, M Tamer},
journal={Internet Computing},
volume={22},
number={5},
pages={18--24},
year={2018},
publisher={IEEE}
}
@article{mhedhbi2019,
author    = {Amine Mhedhbi and
               Semih Salihoglu},
title     = {Optimizing Subgraph Queries by Combining Binary and Worst-Case Optimal
               Joins},
journal   = {CoRR},
volume    = {abs/1903.02076},
year      = {2019},
url       = {http://arxiv.org/abs/1903.02076},
archivePrefix = {arXiv},
eprint    = {1903.02076},
timestamp = {Sun, 31 Mar 2019 19:01:24 +0200},
biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1903-02076},
bibsource = {dblp computer science bibliography, https://dblp.org}
}

@inproceedings{leapfrog1,
title={Adaptive Set Intersections, Unions, and Differences},
author={Demaine, Erik D and L{\'o}pez-Ortiz, Alejandro and Munro, J Ian},
booktitle={Symposium on Discrete Algorithms (SODA)},
year={2000},
organization={ACM-SIAM}
}

@article{leapfrog2,
title={A Simple Algorithm for Merging two Disjoint Linearly Ordered Sets},
author={Hwang, Frank K. and Lin, Shen},
journal={Journal on Computing},
volume={1},
number={1},
pages={31--39},
year={1972},
publisher={SIAM}
}
@inproceedings{uta2019low,
title={low-latency spark queries on updatable data},
author={Uta, Alexandru and Ghit, Bogdan and Dave, Ankur and Boncz, Peter},
booktitle={Proceedings of the 2019 International Conference on Management of Data},
pages={2009--2012},
year={2019},
organization={ACM}
}

@inproceedings{csr-first,
title={Direct Solutions of Sparse Network Equations by Optimally Ordered Triangular Factorization},
author={Tinney, William F and Walker, John W},
booktitle={Proceedings of the IEEE},
volume={55},
number={11},
pages={1801--1809},
year={1967},
publisher={IEEE}
}
@inproceedings{csr,
title={Parallel Sparse Matrix-Vector and Matrix-Transpose-Vector Multiplication using Compressed Sparse Blocks},
author={Bulu{\c{c}}, Aydin and Fineman, Jeremy T and Frigo, Matteo and Gilbert, John R and Leiserson, Charles E},
booktitle={Symposium on Parallelism in Algorithms and Architectures},
pages={233--244},
year={2009},
organization={ACM}
}

@article{fraud-detection,
title={Fraud Detection: Discovering Connections with Graph Databases},
author={Sadowski, Gorka and Rathle, Philip},
journal={White Paper-Neo Technology-Graphs are Everywhere},
year={2014}
}

@inproceedings{fractal,
title={Fractal: A General-Purpose Graph Pattern Mining System},
author={Dias, Vinicius and Teixeira, Carlos HC and Guedes, Dorgival and Meira, Wagner and Parthasarathy, Srinivasan},
booktitle={International Conference on Management of Data (SIGMOD)},
pages={1357--1374},
year={2019},
organization={ACM}
}

@article{longbin,
title={A Survey and Experimental Analysis of Distributed Subgraph Matching},
author={Lai, Longbin and Qing, Zhu and Yang, Zhengyi and Jin, Xin and Lai, Zhengmin and Wang, Ran and Hao, Kongzhang and Lin, Xuemin and Qin, Lu and Zhang, Wenjie and others},
journal={arXiv preprint arXiv:1906.11518},
year={2019}
}

@inproceedings{law,
author = "Paolo Boldi and Sebastiano Vigna",
title = "The {W}eb{G}raph Framework {I}: {C}ompression Techniques",
year = {2004},
booktitle = {World Wide Web Conference (WWW)},
address = {Manhattan, USA},
pages = {595--601},
publisher = {ACM Press}
}

@article{flake2002,
title={Self-Organization and Identification of Web Communities},
author={Flake, Gary William and Lawrence, Steve and Giles, C Lee and Coetzee, Frans M},
journal={Computer},
number={3},
pages={66--71},
year={2002},
publisher={IEEE}
}

@incollection{bodaghi2018automobile,
title={Automobile Insurance Fraud Detection Using Social Network Analysis},
author={Bodaghi, Arezo and Teimourpour, Babak},
booktitle={Applications of Data Management and Analysis},
pages={11--16},
year={2018},
publisher={Springer}
}

@article{newman2004detecting,
title={Detecting Community Structure in Networks},
author={Newman, Mark EJ},
journal={The European Physical Journal B},
volume={38},
number={2},
pages={321--330},
year={2004},
publisher={Springer}
}

@article{neumann2011efficiently,
title={Efficiently Compiling Efficient Query Plans for Modern Hardware},
author={Neumann, Thomas},
journal={VLDB},
volume={4},
number={9},
pages={539--550},
year={2011},
}

@inproceedings{dualsim,
title={DUALSIM: Parallel Subgraph Enumeration in a Massive Graph on a Single Machine},
author={Kim, Hyeonji and Lee, Juneyoung and Bhowmick, Sourav S and Han, Wook-Shin and Lee, JeongHoon and Ko, Seongyun and Jarrah, Moath HA},
booktitle={Conference on Management of Data},
pages={1231--1245},
year={2016},
organization={ACM}
}