Daniel Abadi's Publications

@misc{fabric-vs-mesh,
  author    = {Daniel J. Abadi},
  title     = {Data Fabric vs. Data Mesh: What's the Difference?},
  year      = {2021},
  howpublished = {Starburst Blog. https://blog.starburst.io/data-fabric-vs.-data-mesh-whats-the-difference},
  venue = "Starburst Blog.",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  iis1910613 = "Supported papers",
  url_Paper = "https://blog.starburst.io/data-fabric-vs.-data-mesh-whats-the-difference",
  keywords = "Data mesh,Data fabric",
}

@misc{mesh-answer,
  author    = {Daniel J. Abadi},
  title     = {Data Mesh: The Answer to the Data Warehouse Hypocrisy},
  year      = {2021},
  howpublished = {Starburst Blog. https://blog.starburst.io/data-mesh-the-answer-to-the-data-warehouse-hypocrisy},
  venue = "Starburst Blog.",
  publicationtype = "Blog Post",
  iis1910613 = "Supported papers",
  displayCategory = "Blog Post",
  url_Paper = "https://blog.starburst.io/data-mesh-the-answer-to-the-data-warehouse-hypocrisy",
  keywords = "Data mesh,Analytical database systems",
}

@misc{federation-now-different,
  author    = {Daniel J. Abadi},
  title     = {Data Federation and Data Virtualization Never Worked in the Past But Now it's Different},
  year      = {2021},
  howpublished = {Starburst Blog. https://blog.starburst.io/data-federation-and-data-virtualization-never-worked-in-the-past-but-now-its-different},
  venue = "Starburst Blog.",
  iis1910613 = "Supported papers",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "https://blog.starburst.io/data-federation-and-data-virtualization-never-worked-in-the-past-but-now-its-different",
  keywords = "Data federation,Data virtualization,Analytical database systems",
}

@misc{mesh-integration,
  author    = {Daniel J. Abadi},
  title     = {Does the Data Mesh Make Data Integration Harder?},
  year      = {2021},
  howpublished = {Starburst Blog. https://blog.starburst.io/does-the-data-mesh-make-data-integration-harder},
  venue = "Starburst Blog.",
  iis1910613 = "Supported papers",  
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "https://blog.starburst.io/does-the-data-mesh-make-data-integration-harder",
  keywords = "Data mesh,Data integration,Analytical database systems",
}

@misc{isolation-vs-consistency,
  author    = {Daniel J. Abadi},
  title     = {An explanation of the difference between Isolation levels vs. Consistency levels},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/08/an-explanation-of-difference-between.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/08/an-explanation-of-difference-between.html",
  keywords = "Distributed systems,Scalable transactions,Consistency guarantees",
}

@misc{db-consistency-levels,
  author    = {Daniel J. Abadi},
  title     = {Overview of Consistency Levels in Database Systems},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/07/overview-of-consistency-levels-in.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/07/overview-of-consistency-levels-in.html",
  keywords = "Distributed systems,Consistency guarantees",
}

@misc{dangers-conditional-consistency,
  author    = {Daniel J. Abadi},
  title     = {The dangers of conditional consistency guarantees},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/07/the-dangers-of-conditional-consistency.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/07/the-dangers-of-conditional-consistency.html",
  iis1718581 = "Supported papers",
  keywords = "Distributed systems,PACELC,Consistency guarantees",
}

@misc{correctness-anomalies,
  author    = {Daniel J. Abadi},
  title     = {Correctness Anomalies Under Serializable Isolation},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/06/correctness-anomalies-under.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/06/correctness-anomalies-under.html",
  iis1718581 = "Supported papers",
  keywords = "Distributed systems,Consistency guarantees",
}

@misc{isolation-levels-intro,
  author    = {Daniel J. Abadi},
  title     = {Introduction to Transaction Isolation Levels},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/05/introduction-to-transaction-isolation.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/05/introduction-to-transaction-isolation.html",
  iis1718581 = "Supported papers",
  keywords = "Scalable transactions",
}

@misc{serial-vs-strict-serial,
  author    = {Daniel J. Abadi and Matt Freels},
  title     = {Serializability vs "Strict" Serializability: The Dirty Secret of Database Isolation Levels},
  year      = {2019},
  howpublished = {Fauna Corporate Blog. https://fauna.com/blog/serializability-vs-strict-serializability-the-dirty-secret-of-database-isolation-levels},
  venue = "Fauna Corporate Blog",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "https://fauna.com/blog/serializability-vs-strict-serializability-the-dirty-secret-of-database-isolation-levels",
  keywords = "Distributed systems,Scalable transactions,Consistency guarantees",
}

@misc{move-on-2PC,
  author    = {Daniel J. Abadi},
  title     = {It's Time to Move on from Two Phase Commit},
  year      = {2019},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2019/01/its-time-to-move-on-from-two-phase.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2019/01/its-time-to-move-on-from-two-phase.html",
  iis1527118 = "Supported papers",
  keywords = "Distributed systems,Scalable transactions,Deterministic systems",
}

@misc{consensus-latency,
  author    = {Daniel J. Abadi},
  title     = {Partitioned consensus and its impact on Spanner's latency},
  year      = {2018},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2018/12/partitioned-consensus-and-its-impact-on.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2018/12/partitioned-consensus-and-its-impact-on.html",
  iis1527118 = "Supported papers",
  keywords = "Distributed systems,Scalable transactions,Deterministic systems",
}

@misc{newsql-failure,
  author    = {Daniel J. Abadi},
  title     = {NewSQL database systems are failing to guarantee consistency, and I blame Spanner},
  year      = {2018},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2018/09/newsql-database-systems-are-failing-to.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2018/09/newsql-database-systems-are-failing-to.html",
  iis1527118 = "Supported papers",
  keywords = "Distributed systems,Scalable transactions,Deterministic systems",
}

@misc{arrow-strengths,
  author    = {Daniel J. Abadi},
  title     = {An analysis of the strengths and weaknesses of Apache Arrow},
  year      = {2018},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2018/03/an-analysis-of-strengths-and-weaknesses.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2018/03/an-analysis-of-strengths-and-weaknesses.html",
  keywords = "Analytical database systems,Column-stores",
}

@misc{arrow-parquet-orc,
  author    = {Daniel J. Abadi},
  title     = {Apache Arrow vs. Parquet and ORC: Do we really need a third Apache project for columnar data representation?},
  year      = {2017},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2017/10/apache-arrow-vs-parquet-and-orc-do-we.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2017/10/apache-arrow-vs-parquet-and-orc-do-we.html",
  keywords = "Analytical database systems,Column-stores",
}

@misc{hazelcast-pacelc,
  author    = {Daniel J. Abadi},
  title     = {Hazelcast and the Mythical PA/EC System},
  year      = {2017},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2017/10/hazelcast-and-mythical-paec-system.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2017/10/hazelcast-and-mythical-paec-system.html",
  keywords = "Distributed systems,PACELC,Consistency guarantees",
}

@misc{spanner-vs-calvin,
  author    = {Daniel J. Abadi},
  title     = {Distributed consistency at scale: Spanner vs. Calvin},
  year      = {2017},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2017/04/distributed-consistency-at-scale.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2017/04/distributed-consistency-at-scale.html",
  keywords = "Distributed systems,Scalable transactions,Deterministic systems,Calvin",
}

@misc{single-item-xacts,
  author    = {Daniel J. Abadi and Jose Faleiro},
  title     = {Why MongoDB, Cassandra, HBase, DynamoDB, and Riak will only let you perform transactions on a single data item},
  year      = {2015},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2015/10/why-mongodb-cassandra-hbase-dynamodb_28.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2015/10/why-mongodb-cassandra-hbase-dynamodb_28.html",
  keywords = "Distributed systems,Scalable transactions",
}

@misc{acid-problems,
  author    = {Daniel J. Abadi and Alexander Thomson},
  title     = {The problems with ACID, and how to fix them without going NoSQL},
  year      = {2010},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2010/08/problems-with-acid-and-how-to-fix-them.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2010/08/problems-with-acid-and-how-to-fix-them.html",
  keywords = "Distributed systems,Scalable transactions,Deterministic systems,Calvin",
}

@misc{pacelc-blog,
  author    = {Daniel J. Abadi},
  title     = {Problems with CAP, and Yahoo's little known NoSQL system},
  year      = {2010},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2010/04/problems-with-cap-and-yahoos-little.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2010/04/problems-with-cap-and-yahoos-little.html",
  keywords = "Distributed systems,PACELC,Consistency guarantees",
}

@misc{two-types-column-stores,
  author    = {Daniel J. Abadi},
  title     = {Distinguishing Two Major Types of Column-Stores},
  year      = {2010},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2010/03/distinguishing-two-major-types-of_29.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2010/03/distinguishing-two-major-types-of_29.html",
  keywords = "Analytical database systems,Column-stores",
}

@misc{hybrid-column-tour,
  author    = {Daniel J. Abadi},
  title     = {A tour through hybrid column/row-oriented DBMS schemes},
  year      = {2009},
  howpublished = {DBMS Musings. http://dbmsmusings.blogspot.com/2009/09/tour-through-hybrid-columnrow-oriented.html},
  venue = "DBMS Musings",
  publicationtype = "Blog Post",
  displayCategory = "Blog Post",
  url_Paper = "http://dbmsmusings.blogspot.com/2009/09/tour-through-hybrid-columnrow-oriented.html",
  keywords = "Analytical database systems,Column-stores",
}

%% Demonstrations

@misc{c5,
  author    = {Jeffrey Helt and
               Abhinav Sharma and
               Daniel J. Abadi and
               Wyatt Lloyd and
               Jose M. Faleiro},
  title     = {{C5:} Cloned Concurrency Control That Always Keeps Up},
  journal   = {Proc. {VLDB} Endow.},
  volume    = {16},
  number    = {1},
  pages     = {1--14},
  year      = {2022},
  howpublished = {PVLDB 16(1)},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/c5-paper.pdf",
}

@inproceedings{bullfrog,
        title = {BullFrog: Online Schema Evolution via Lazy Evaluation},
        year = "2021",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Souvik Bhattacherjee and Gang Liao and Michael Hicks and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/bullfrog-sigmod.pdf",
	pdfKB = "1290",
        iis1910613 = "Supported papers",
        iis1718581 = "Supported papers",
        keywords = "Schema evolution, Schema management, Scalable transactions",
}

@inproceedings{anylog,
  author    = {Daniel J. Abadi and Owen Arden and Faisal Nawab and Moshe Shadmon},
  title     = {AnyLog: a Grand Unification of the Internet of Things},
  booktitle = "CIDR",
  year      = "2020",
  venue = "CIDR",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/anylogAbadiEtAl.pdf",
  pdfKB = "256",
  iis1910613 = "Supported papers",
  keywords = "Decentralized database systems; Peer-to-peer database systems; Blockchain",
}

@misc{hadoopdb-retrospective,
  author    = {Azza Abouzied and Daniel J. Abadi and Kamil Bajda-Pawlikowski and Avi Silberschatz},
  title     = {Integration of Large-Scale Data Processing Systems and Traditional Parallel Database Technology},
  journal   = {{PVLDB}},
  volume    = {12},
  number    = {12},
  pages     = {2290--2299},
  year      = {2019},
  howpublished = {PVLDB 12(12)},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/20001-Abadi.pdf",
  pdfKB = "",
  iis1910613 = "Supported papers",
  iis1718581 = "Supported papers",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce,Non-relational data,Schema management,Graph database systems",
  project:multiple = "HadoopDB",
}

@misc{slog,
  author    = {Kun Ren and Dennis Li and Daniel J. Abadi},
  title     = {SLOG: Serializable, Low-latency, Geo-replicated Transactions},
  journal   = {{PVLDB}},
  volume    = {12},
  number    = {11},
  pages     = {1747--1761},
  year      = {2019},
  howpublished = {PVLDB 12(11)},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/1154-Abadi.pdf",
  pdfKB = "",
  iis1527118 = "Supported papers",
  iis1718581 = "Supported papers",
  keywords = "Scalable transactions,Deterministic systems,Distributed systems,Calvin",
  project:multiple = "SLOG;Calvin",
}

@inproceedings {fuzzylog,
  author = {Joshua Lockerman and Jose M. Faleiro and Juno Kim and Soham Sankaran and Daniel J. Abadi and James Aspnes and Siddhartha Sen and Mahesh Balakrishnan},
  title = {The FuzzyLog: A Partially Ordered Shared Log},
  booktitle = "OSDI",
  year      = "2018",
  venue = "OSDI",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/fuzzylog.pdf",
  pdfKB = "7244",
  iis1718581 = "Supported papers",
  keywords = "Scalable transactions,Distributed systems",
  project:multiple = "SLOG;Calvin",
}

@misc{determinism-cacm,
  author = {Daniel J. Abadi and Jose M. Faleiro},
  title = {An Overview of Deterministic Database Systems},
  howpublished = {CACM, 61(9)},
  month = {September},
  year = {2018},
  pages = {78-88},
  venue = {CACM},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-cacm2018.pdf",
  pdfKB = "5530",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  iis1527118 = "Supported papers",
  iis1718581 = "Supported papers",
  keywords = "Calvin,Scalable transactions,Distributed systems,Deterministic systems,Main-memory database systems",
  project:multiple = "SLOG;Calvin",
}

@misc{early-write-visibility,
  author    = {Jose M. Faleiro and Daniel J. Abadi and Joseph M. Hellerstein},
  title     = {High Performance Transactions via Early Write Visibility},
  journal   = {{PVLDB}},
  volume    = {10},
  number    = {5},
  pages     = {613-624},
  year      = {2017},
  howpublished = {PVLDB 10(5): 613-624},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/early-write-visibility.pdf",
  pdfKB = "269",
  iis1527118 = "Supported papers",
  iis1718581 = "Relevant prior work",
  keywords = "Scalable transactions,Deterministic systems,Multi-core systems,Main-memory database systems",
  project:multiple = "SLOG",
}

@inproceedings{latch-free-sync,
  author    = {Jose M. Faleiro and Daniel J. Abadi},
  title     = {Latch-free Synchronization in Database Systems: Silver Bullet or Fool's Gold?},
  booktitle = "CIDR",
  year      = "2017",
  venue = "CIDR",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/latch-free-cidr2017.pdf",
  pdfKB = "378",
  iis1527118 = "Supported papers",
  keywords = "Scalable transactions,Multi-core systems,Main-memory database systems",
}

@inproceedings{schemagen,
        title = {Automatic Generation of Normalized Relational Schemas from Nested Key-Value Data},
        year = "2016",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Michael DiScala and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/schemagen-sigmod16.pdf",
	pdfKB = "926",
        iis1527118 = "Supported papers",
        keywords = "Non-relational data, Schema management",
}

@inproceedings{orthrus,
        title = {Design Principles for Scaling Multi-core OLTP Under High Contention},
        year = "2016",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Kun Ren and Jose Faleiro and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/orthrus-sigmod16.pdf",
	pdfKB = "349",
        iis1527118 = "Supported papers",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems",
}

@inproceedings{fast-checkpoint,
        title = {Low-Overhead Asynchronous Checkpointing in Main-Memory Database Systems},
        year = "2016",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Kun Ren and Thaddeus Diamond and Daniel J. Abadi and Alexander Thomson},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/fast-checkpoint-sigmod16.pdf",
	pdfKB = "436",
        iis1527118 = "Supported papers",
        keywords = "Scalable transactions,Main-memory database systems,Deterministic systems",
}

@inproceedings{graph-dedense,
        title = {Scalable Pattern Matching Over Compressed Graphs via Dedensification},
        year = "2016",
        booktitle = "KDD",
        venue = "KDD",
        author = {Antonio Maccioni and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/graph-dedense.pdf",
	pdfKB = "678",
        iis1527118 = "Supported papers",
        keywords = "Non-relational data, Graph database systems",
}

@misc{leopard,
  author    = {Jiewen Huang and Daniel J. Abadi},
  title     = {LEOPARD: Lightweight Edge-Oriented Partitioning and Replication for Dynamic Graphs},
  journal   = {{PVLDB}},
  volume    = {9},
  number    = {7},
  pages     = {540-551},
  year      = {2016},
  howpublished = {PVLDB 9(7): 540-551},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/leopard-vldb16.pdf",
  pdfKB = "609",
  iis1527118 = "Supported papers",
  keywords = "Non-relational data, Graph database systems,Distributed systems",
}

@misc{rethink-mvcc,
  author    = {Jose M. Faleiro and Daniel J. Abadi},
  title     = {Rethinking serializable multiversion concurrency control},
  journal   = {{PVLDB}},
  volume    = {8},
  number    = {11},
  pages     = {1190-1201},
  year      = {2015},
  howpublished = {PVLDB 8(11): 1190-1201},
  venue = "PVLDB",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/rethink-mvcc.pdf",
  pdfKB = "524",
  iis1527118 = "Relevant prior work",
  iis1718581 = "Relevant prior work",
  iis1249722 = "Supported papers",
  keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Deterministic systems",
}

@inproceedings{calvin-fs,
  author    = {Alexander Thomson and Daniel J. Abadi},
  title     = {CalvinFS: Consistent WAN Replication and Scalable Metadata Management for Distributed File Systems},
  booktitle = {FAST},
  venue = "FAST",
  pages     = {1-14},
  year      = {2015},
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/calvinfs.pdf",
  pdfKB = "933",
  keywords = "Scalable transactions,Deterministic systems,File systems,Distributed systems,Calvin",
  project:multiple = "Calvin",
}

@misc{fit,
  author    = {Jose M. Faleiro and Daniel J. Abadi},
  title     = {FIT: A Distributed Database Performance Tradeoff},
  journal   = {{IEEE} Data Eng. Bull.},
  volume    = {38},
  number    = {1},
  pages     = {10-17},
  year      = {2015},
  howpublished   = {IEEE Data Engineering Bulletin, 38(1): 10-17},
  venue   = {IEEE Data Engineering Bulletin},
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/fit.pdf",
  pdfKB = "51",
  keywords = "Scalable transactions,Distributed systems",
}

@misc{vll-vldbj,
  author = {Kun Ren and Alexander Thomson and Daniel J. Abadi},
  title = {VLL: A Lock Manager Redesign for Main Memory Database Systems},
  journal = {{VLDB Journal}},
  volume = {24},
  number = {5},
  howpublished = {VLDB Journal 24(5): 681-705},
  month = {October},
  year = {2015},
  pages = {681-705},
  venue = {VLDB Journal},
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldbj-vll.pdf",
  pdfKB = "3464",
  publicationtype = "Journal Article",
  iis1527118 = "Relevant prior work",
  iis1249722 = "Supported papers",
  keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Distributed systems,Calvin,Deterministic systems",
  project:multiple = "Calvin",
}

@inproceedings{lazy-xacts,
        title = "Lazy Evaluation of Transactions in Database Systems",
        year = "2014",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Jose Faleiro and Alexander Thomson and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/lazy-xacts.pdf",
	pdfKB = "774",
        iis1527118 = "Relevant prior work",
        iis1718581 = "Relevant prior work",
        iis1249722 = "Supported papers",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Deterministic systems",
}

@inproceedings{sinew,
        title = "Sinew: a SQL System for Unified Analytics of Multi-structured Data",
        year = "2014",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Daniel Tahara and Thaddeus Diamond and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/sinew-sigmod14.pdf",
	pdfKB = "715",
        iis0845643 = "Supported papers",
        keywords = "Non-relational data, Schema management",
}

@misc{determinism-eval,
        title = "An Evaluation of the Advantages and Disadvantages of Deterministic Database Systems",
        howpublished = {PVLDB 7(10): 821-832},
        year = "2014",
        venue = "PVLDB",
        author = {Kun Ren and Alexander Thomson and Daniel J. Abadi},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/determinism-eval.pdf",
	pdfKB = "524",
        iis1527118 = "Relevant prior work",
        iis1249722 = "Supported papers",
        keywords = "Scalable transactions,Distributed systems,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@misc{calvin-tods,
        title = "Fast Distributed Transactions and Strongly Consistent Replication for OLTP Database Systems",
        howpublished = {ACM TODS 39(2): 11},
        year = "2014",
        venue = "TODS",
        author = {Alexander Thomson and Thaddeus Diamond and Shu-Chun Weng and Kun Ren and Philip Shao and Daniel J. Abadi},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/calvin-tods14.pdf",
	pdfKB = "1987",
        keywords = "Scalable transactions,Distributed systems,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@inproceedings{subgraph-opt,
        title = "Query Optimization of Distributed Pattern Matching",
        year = "2014",
        booktitle = "ICDE",
        venue = "ICDE",
        author = {Jiewen Huang and Kartik Venkatraman and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/subgraph-opt.pdf",
	pdfKB = "446",
        iis0845643 = "Supported papers",
        keywords = "Non-relational data,Graph database systems,Distributed systems,Scalable queries,Parallel database systems",
}

@misc{beckman-report,
  author    = {Daniel J. Abadi and Rakesh Agrawal and Anastasia Ailamaki and Magdalena Balazinska and Philip A. Bernstein and Michael J. Carey and Surajit Chaudhuri and Jeffrey Dean and AnHai Doan and Michael J. Franklin and Johannes Gehrke and Laura M. Haas and Alon Y. Halevy and Joseph M. Hellerstein and Yannis E. Ioannidis and H. V. Jagadish and Donald Kossmann and Samuel Madden and Sharad Mehrotra and Tova Milo and Jeffrey F. Naughton and Raghu Ramakrishnan and Volker Markl and Christopher Olston and Beng Chin Ooi and Christopher Re and Dan Suciu and Michael Stonebraker and Todd Walter and Jennifer Widom},
  title     = {The Beckman Report on Database Research},
  journal   = {{SIGMOD} Record},
  volume    = {43},
  number    = {3},
  pages     = {61-70},
  year      = {2014},
  howpublished   = {SIGMOD Record, 43(3): 61-70},
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/beckman-report2013.pdf",
  pdfKB = "122",
}

@misc{modular-calvin,
  author    = {Alexander Thomson and Daniel J. Abadi},
  title     = {Modularity and Scalability in Calvin},
  howpublished   = {IEEE Data Engineering Bulletin, 36(2): 48-55},
  venue   = {IEEE Data Engineering Bulletin},
  volume    = {36},
  number    = {2},
  year      = {2013},
  pages     = {48-55},
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/scalable-calvin.pdf",
  pdfKB = "628",
  iis1527118 = "Relevant prior work",
  iis1249722 = "Supported papers",
  keywords = "Scalable transactions,Distributed systems,Calvin,Deterministic systems",
  project:multiple = "Calvin",
}

@misc{abadi-column-store-survery,
        number = {3},
        volume = {5},
        pages = {197-280},
        year = {2013},
        title = {The Design and Implementation of Modern Column-Oriented Database Systems},
        author = {Daniel Abadi and Peter Boncz and Stavros Harizopoulos and Stratos Idreos and Samuel Madden},
        howpublished = {Foundations and Trends in Databases, 5(3): 197-280},
        venue = {Foundations and Trends in Databases},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-column-stores.pdf",
	pdfKB = "1997",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store",
        project:multiple = "C-Store",
}

@inproceedings{invisible-loading,
        title = "Invisible Loading: Access-Driven Data Transfer from Raw Files into Database Systems",
        year = "2013",
        booktitle = "EDBT",
        venue = "EDBT",
        author = {Azza Abouzied and Daniel J. Abadi and Avi Silberschatz},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/invisibleloading.pdf",
	pdfKB = "470",
        iis0845643 = "Supported papers",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce",
        project:multiple = "HadoopDB",
}

@misc{abadi-vll,
        title = "Lightweight Locking for Main Memory Database Systems",
        howpublished = {PVLDB 6(2): 145-156},
        year = "2012",
        venue = "PVLDB",
        author = {Kun Ren and Alexander Thomson and Daniel J. Abadi},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/vll-vldb13.pdf",
	pdfKB = "1018",
        iis1249722 = "Supported papers",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Distributed systems,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@misc{abadi-pacelc,
  author = {Daniel J. Abadi},
  title = {Consistency Tradeoffs in Modern Distributed Database System Design: CAP is Only Part of the Story},
  howpublished = {IEEE Computer, 45(2)},
  year = {2012},
  pages = {37-42},
  venue = {IEEE Computer},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-pacelc.pdf",
  pdfKB = "62",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  keywords = "Scalable transactions,Distributed systems,Consistency guarantees,PACELC",
}

@inproceedings{calvin,
        title = "Calvin: Fast Distributed Transactions for Partitioned Database Systems",
        year = "2012",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Alexander Thomson and Thaddeus Diamond and Shu-Chun Weng and Kun Ren and Philip Shao and Daniel J. Abadi},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/calvin-sigmod12.pdf",
	pdfKB = "544",
        iis1527118 = "Relevant prior work",
        iis0845643 = "Supported papers",
        keywords = "Scalable transactions,Distributed systems,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@misc{scalable-subgraphs,
  author = {Jiewen Huang and Daniel J. Abadi and Kun Ren},
  title = {Scalable SPARQL Querying of Large RDF Graphs},
  howpublished = {PVLDB, 4(21)},
  month = {August},
  year = {2011},
  pages = {1123-1134},
  venue = {PVLDB},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/sw-graph-scale.pdf",
  pdfKB = "457",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  iis0845643 = "Supported papers",
  keywords = "Non-relational data, Graph database systems, Distributed systems,Scalable queries,Parallel database systems,Analytical database systems",
}

@inproceedings{split-execution,
        title = "Efficient Processing of Data Warehousing Queries in a Split Execution Environment",
        year = "2011",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Kamil Bajda-Pawlikowski and Daniel J. Abadi and Avi Silberschatz and Erik Paulson},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/split-execution-hadoopdb.pdf",
	pdfKB = "609",
        nsfclue = "Supported papers",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce",
        project:multiple = "HadoopDB",
}

@inproceedings{transactions-wodet11,
        title = "Building Deterministic Transaction Processing Systems without Deterministic Thread Scheduling",
        year = "2011",
        booktitle = "WODET",
        venue = "The 2nd Workshop on Determinism and Correctness in Parallel Programming",
        author = {Alexander Thomson and Daniel J. Abadi},
	publicationtype = "Workshop Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/transactions-wodet11.pdf",
	pdfKB = "223",
        keywords = "Scalable transactions,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@misc{dbms-determinism,
        title = "The Case for Determinism in Database Systems",
        howpublished = {PVLDB, 3(1)},
        month = {September},
        year = "2010",
        venue = "PVLDB",
        pages = {70-80},
        author = {Alexander Thomson and Daniel J. Abadi},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/determinism-vldb10.pdf",
	pdfKB = "550",
        iis0845643 = "Supported papers",
        keywords = "Scalable transactions,Distributed systems,Calvin,Deterministic systems",
        project:multiple = "Calvin",
}

@inproceedings{hstore-cc,
        title = "Low Overhead Concurrency Control for Partitioned Main Memory Databases",
        year = "2010",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        author = {Evan P. C. Jones and Daniel J. Abadi and Samuel Madden},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/hstore-cc.pdf",
	pdfKB = "300",
        iis0845643 = "Supported papers",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Distributed systems,H-Store",
        project:multiple = "H-Store",
}

@misc{mr-cacm,
  author = {Michael Stonebraker and Daniel J. Abadi and David. J. DeWitt and Samuel Madden and Erik Paulson and Andrew Pavlo and Alexander Rasin},
  title = {MapReduce and Parallel DBMSs: Friends or Foes?},
  howpublished = {CACM, 53(1)},
  month = {January},
  year = {2010},
  pages = {64-71},
  venue = {CACM},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/mr-cacm2010.pdf",
  pdfKB = "3625",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  nsfclue = "Supported papers",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,MapReduce",
}

@misc{hadoopdb,
        title = "HadoopDB: An Architectural Hybrid of MapReduce and DBMS Technologies for Analytical Workloads",
        howpublished = {PVLDB, 2(1)},
        month = {August},
        year = {2009},
        pages = {922-933},
        venue = {PVLDB},
        author = {Azza Abouzied and Kamil Bajda-Pawlikowski and Daniel J. Abadi and Avi Silberschatz and Alexander Rasin},
	publicationtype = "Journal Article",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/hadoopdb.pdf",
	pdfKB = "400",
        iis0845643 = "Supported papers",
        award = "VLDB Test of Time Award",
        note = "<u>VLDB Test of Time Award</u>",
        nsfclue = "Supported papers",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce",
        project:multiple = "HadoopDB",
}

@inproceedings{benchmarks-sigmod09,
        title = "A Comparison of Approaches to Large Scale Data Analysis",
        year = "2009",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        address = "Providence, Rhode Island, USA",
        author = {Andrew Pavlo and Erik Paulson and Alexander Rasin and Daniel J. Abadi and David J. DeWitt and Samuel R. Madden and Michael Stonebraker},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/benchmarks-sigmod09.pdf",
	pdfKB = "251",
        nsfclue = "Supported papers",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,MapReduce",
}

@misc{abadi-ieee-cloud,
  author = {Daniel J. Abadi},
  title = {Data Management in the Cloud: Limitations and Opportunities},
  howpublished = {IEEE Data Engineering Bulletin, 32(1)},
  month = {March},
  year = {2009},
  pages = {3-12},
  venue = {IEEE Data Engineering Bulletin},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-cloud-ieee09.pdf",
  pdfKB = "62",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  iis0845643 = "Supported papers",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,MapReduce",
}

@misc{abadi-swstore,
  author = {Daniel J. Abadi and Adam Marcus and Samuel R. Madden and Kate Hollenbach},
  title = {SW-Store: A Vertically Partitioned DBMS for Semantic Web Data Management},
  howpublished = {VLDB Journal, 18(2)},
  month = {April},
  year = {2009},
  pages = {385-406},
  venue = {VLDB Journal},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-rdf-vldbj.pdf",
  pdfKB = "1124",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  iis0845643 = "Supported papers",
  keywords = "Non-relational data,Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
}

@inproceedings{abadi-sigmod08,
        title = "Column-Stores vs. Row-Stores: How Different Are They Really?",
        year = "2008",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        address = "Vancouver, Canada",
        author = {Daniel J. Abadi and Samuel R. Madden and Nabil Hachem},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadi-sigmod08.pdf",
	pdfKB = "424",
        keywords = "Analytical database systems,Column-stores,C-Store",
        project:multiple = "C-Store",
}

@inproceedings{oltp-perf,
        title = "OLTP Through the Looking Glass, And What We Found There",
        year = "2008",
        booktitle = "SIGMOD",
        venue = "SIGMOD",
        address = "Vancouver, Canada",
        author = {Stavros Harizopoulos and Daniel J. Abadi and Samuel R. Madden and Michael Stonebraker},
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
	pdfKB = "287",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/oltpperf-sigmod08.pdf",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,H-Store",
        project:multiple = "H-Store",
}

@inproceedings{abadi-rdf,
	title = "Scalable Semantic Web Data Management Using Vertical Partitioning",
	year = "2007",
        booktitle = "VLDB",
        address = "Vienna, Austria",
	venue = "VLDB",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadirdf.pdf",
	author = {Daniel J. Abadi and Adam Marcus and Samuel R. Madden and Kate Hollenbach},
	abstract = "Efficient management of RDF data is an important factor in realizing the Semantic Web vision. Performance and scalability issues are becoming increasingly pressing as Semantic Web technology is applied to real-world applications. In this paper, we examine the reasons why current data management solutions for RDF data scale poorly, and explore the fundamental scalability limitations of these approaches. We review the state of the art for improving performance for RDF databases and consider a recent suggestion, 'property tables'. We then discuss practically and empirically why this solution has undesirable features. As an improvement, we propose an alternative solution: vertically partitioning the RDF data. We compare the performance of vertical partitioning with prior art on queries generated by a Web-based RDF browser over a large-scale (more than 50 million triples) catalog of library data.  Our results show that a vertical partitioned schema achieves similar performance to the property table technique while being much simpler to design. Further, if a column-oriented DBMS (a database architected specially for the vertically partitioned case) is used instead of a row-oriented DBMS, another order of magnitude performance improvement is observed, with query times dropping from minutes to several seconds.",
	pdfKB = "246",
	publicationtype = "Conference Paper",
        displayCategory = "Conference or Journal Publication",
        award = "Best Paper Award",
        note = "<u>Best Paper Award</u>",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store,Non-relational data",
        project:multiple = "C-Store",
}

Efficient management of RDF data is an important factor in realizing the Semantic Web vision. Performance and scalability issues are becoming increasingly pressing as Semantic Web technology is applied to real-world applications. In this paper, we examine the reasons why current data management solutions for RDF data scale poorly, and explore the fundamental scalability limitations of these approaches. We review the state of the art for improving performance for RDF databases and consider a recent suggestion, 'property tables'. We then discuss practically and empirically why this solution has undesirable features. As an improvement, we propose an alternative solution: vertically partitioning the RDF data. We compare the performance of vertical partitioning with prior art on queries generated by a Web-based RDF browser over a large-scale (more than 50 million triples) catalog of library data. Our results show that a vertical partitioned schema achieves similar performance to the property table technique while being much simpler to design. Further, if a column-oriented DBMS (a database architected specially for the vertically partitioned case) is used instead of a row-oriented DBMS, another order of magnitude performance improvement is observed, with query times dropping from minutes to several seconds.

@inproceedings{hstore,
  author    = {Michael Stonebraker and Samuel R. Madden and Daniel J. Abadi and Stavros Harizopoulos and Nabil Hachem and Pat Helland},
  title     = {The End of an Architectural Era (It's Time for a Complete Rewrite)},
  booktitle = {VLDB},
  year      = {2007},
  address = {Vienna, Austria},
  venue = "VLDB",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldb07hstore.pdf",
  abstract = "In previous papers, some of us predicted the end of 'one size fits all' as a commercial relational DBMS paradigm. These papers presented reasons and experimental evidence that showed that the major relational RDBMS vendors can be outperformed by 1-2 orders of magnitude by specialized engines in the data warehouse, stream processing, text, and scientific database markets. Assuming that specialized engines dominate these markets over time, the current relational DBMS code lines will be left with the business data processing (OLTP) market and hybrid markets where more than one capability is required. In this paper we show that current RDBMSs can be beaten by nearly two orders of magnitude in the OLTP market as well. The experimental evidence comes from comparing a new OLTP prototype, H-Store, which we have built at M.I.T., to a popular RDBMS on the standard transactional benchmark, TPC-C. We conclude that current RDBMS code lines, while attempting to be a 'one size fits all' solution, in fact, excel at nothing. Hence, they are 25 year old legacy code lines that should be retired in favor of a collection of 'from scratch' specialized engines. The DBMS vendors (and research community) should start with a clean sheet of paper and design systems for tomorrow's requirements, not continue to push code lines and architectures designed for yesterday's needs.",
  pdfKB = "444",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Distributed systems,H-Store",
  project:multiple = "H-Store",
}

In previous papers, some of us predicted the end of 'one size fits all' as a commercial relational DBMS paradigm. These papers presented reasons and experimental evidence that showed that the major relational RDBMS vendors can be outperformed by 1-2 orders of magnitude by specialized engines in the data warehouse, stream processing, text, and scientific database markets. Assuming that specialized engines dominate these markets over time, the current relational DBMS code lines will be left with the business data processing (OLTP) market and hybrid markets where more than one capability is required. In this paper we show that current RDBMSs can be beaten by nearly two orders of magnitude in the OLTP market as well. The experimental evidence comes from comparing a new OLTP prototype, H-Store, which we have built at M.I.T., to a popular RDBMS on the standard transactional benchmark, TPC-C. We conclude that current RDBMS code lines, while attempting to be a 'one size fits all' solution, in fact, excel at nothing. Hence, they are 25 year old legacy code lines that should be retired in favor of a collection of 'from scratch' specialized engines. The DBMS vendors (and research community) should start with a clean sheet of paper and design systems for tomorrow's requirements, not continue to push code lines and architectures designed for yesterday's needs.

@inproceedings{abadi-cidr,
  author    = {Daniel J. Abadi},
  title     = {Column Stores for Wide and Sparse Data},
  booktitle = {CIDR},
  year      = {2007},
  address  = {Asilomar, CA, USA},
  venue = "CIDR",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadicidr07.pdf",
  abstract = "While it is generally accepted that data warehouses and OLAP workloads are excellent applications for column-stores, this paper speculates that column-stores may well be suited for additional applications. In particular we observe that column-stores do not see a performance degradation when storing extremely wide tables, and column-stores handle sparse data very well. These two properties lead us to conjecture that column-stores may be good storage layers for Semantic Web data, XML data, and data with GEM-style schemas.",
  pdfKB = "156",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Column-stores,C-Store",
  project:multiple = "C-Store",
}

While it is generally accepted that data warehouses and OLAP workloads are excellent applications for column-stores, this paper speculates that column-stores may well be suited for additional applications. In particular we observe that column-stores do not see a performance degradation when storing extremely wide tables, and column-stores handle sparse data very well. These two properties lead us to conjecture that column-stores may be good storage layers for Semantic Web data, XML data, and data with GEM-style schemas.

@inproceedings{cstore-mat,
  author    = {Daniel J. Abadi and Daniel S. Myers and David J. DeWitt and Samuel R. Madden},
  title     = {Materialization Strategies in a Column-Oriented DBMS},
  booktitle = {ICDE},
  year      = {2007},
  address  = {Istanbul, Turkey},
  pages = {466-475},
  venue = "ICDE",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadiicde2007.pdf",
  abstract = "There has been renewed interest in column-oriented database architectures in recent years. For read-mostly query workloads such as those found in data warehouse and decision support applications, column-stores have been shown to perform particularly well relative to row-stores. In order for column-stores to be readily adopted as a replacement for row-stores, however, they must present the same interface to client applications as do row stores, which implies that they must output row-store-style tuples. Thus, the input columns stored on disk must be converted to rows at some point in the query plan, but the optimal point at which to do the conversion is not obvious. This problem can be considered as the opposite of the projection problem in row-store systems: while row-stores need to determine where in query plans to place projection operators to make tuples narrower, column-stores need to determine when to combine single-column projections into wider tuples. This paper describes a variety of strategies for tuple construction and intermediate result representations and provides a systematic evaluation of these strategies.",
  pdfKB = "327",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
}

There has been renewed interest in column-oriented database architectures in recent years. For read-mostly query workloads such as those found in data warehouse and decision support applications, column-stores have been shown to perform particularly well relative to row-stores. In order for column-stores to be readily adopted as a replacement for row-stores, however, they must present the same interface to client applications as do row stores, which implies that they must output row-store-style tuples. Thus, the input columns stored on disk must be converted to rows at some point in the query plan, but the optimal point at which to do the conversion is not obvious. This problem can be considered as the opposite of the projection problem in row-store systems: while row-stores need to determine where in query plans to place projection operators to make tuples narrower, column-stores need to determine when to combine single-column projections into wider tuples. This paper describes a variety of strategies for tuple construction and intermediate result representations and provides a systematic evaluation of these strategies.

@inproceedings{cstore-perf,
  author = {Stavros Harizopoulos and Velen Liang and Daniel J. Abadi and Samuel R. Madden},
  title = {Performance Tradeoffs in Read-Optimized Databases},
  booktitle = {VLDB},
  year = {2006},
  pages = {487-498},
  address = {Seoul, Korea},
  venue = "VLDB",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/VLDB06.pdf",
  abstract = "Database systems have traditionally optimized performance for write-intensive workloads. Recently, there has been renewed interest in architectures that optimize read performance by using column-oriented data representation and light-weight compression. This previous work has shown that under certain broad classes of workloads, column-based systems can outperform row-based systems. Previous work, however, has not characterized the precise conditions under which a particular query workload can be expected to perform better on a column-oriented database. In this paper we first identify the distinctive components of a read-optimized DBMS and describe our implementation of a high-performance query engine that can operate on both row and column-oriented data. We then use our prototype to perform an in-depth analysis of the tradeoffs between column and row-oriented architectures. We explore these tradeoffs in terms of disk bandwidth, CPU cache latency, and CPU cycles. We show that for most database workloads, a carefully designed column system can outperform a carefully designed row system, sometimes by an order of magnitude. We also present an analytical model to predict whether a given workload on a particular hardware configuration is likely to perform better on a row or column-based system.",
  pdfKB = "354",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
 }

Database systems have traditionally optimized performance for write-intensive workloads. Recently, there has been renewed interest in architectures that optimize read performance by using column-oriented data representation and light-weight compression. This previous work has shown that under certain broad classes of workloads, column-based systems can outperform row-based systems. Previous work, however, has not characterized the precise conditions under which a particular query workload can be expected to perform better on a column-oriented database. In this paper we first identify the distinctive components of a read-optimized DBMS and describe our implementation of a high-performance query engine that can operate on both row and column-oriented data. We then use our prototype to perform an in-depth analysis of the tradeoffs between column and row-oriented architectures. We explore these tradeoffs in terms of disk bandwidth, CPU cache latency, and CPU cycles. We show that for most database workloads, a carefully designed column system can outperform a carefully designed row system, sometimes by an order of magnitude. We also present an analytical model to predict whether a given workload on a particular hardware configuration is likely to perform better on a row or column-based system.

@inproceedings{cstore-comp,
  author    = {Daniel J. Abadi and Samuel R. Madden and Miguel Ferreira},
  title     = {Integrating Compression and Execution in Column-Oriented Database Systems},
  booktitle = {SIGMOD},
  year      = {2006},
  address   = {Chicago, IL, USA},
  pages     = {671-682},
  venue = "SIGMOD",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadisigmod06.pdf",
  abstract = "Column-oriented database system architectures invite a re-evaluation of how and when data in databases is compressed. Storing data in a column-oriented fashion greatly increases the similarity of adjacent records on disk and thus opportunities for compression. The ability to compress many adjacent tuples at once lowers the per-tuple cost of compression, both in terms of CPU and space overheads. In this paper, we discuss how we extended C-Store (a column-oriented DBMS) with a compression sub-system. We show how compression schemes not traditionally used in row-oriented DBMSs can be applied to column-oriented systems.  We then evaluate a set of compression schemes and show that the best scheme depends not only on the properties of the data but also on the nature of the query workload.",
  pdfKB = "265",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Scalable queries,Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
}

Column-oriented database system architectures invite a re-evaluation of how and when data in databases is compressed. Storing data in a column-oriented fashion greatly increases the similarity of adjacent records on disk and thus opportunities for compression. The ability to compress many adjacent tuples at once lowers the per-tuple cost of compression, both in terms of CPU and space overheads. In this paper, we discuss how we extended C-Store (a column-oriented DBMS) with a compression sub-system. We show how compression schemes not traditionally used in row-oriented DBMSs can be applied to column-oriented systems. We then evaluate a set of compression schemes and show that the best scheme depends not only on the properties of the data but also on the nature of the query workload.

@inproceedings{reed,
  author    = {Daniel J. Abadi and Samuel R. Madden and Wolfgang Lindner},
  title     = {REED: Robust, Efficient Filtering and Event Detection in Sensor Networks},
  booktitle = {VLDB},
  year      = {2005},
  address   = {Trondheim, Norway},
  pages     = {769-780},
  venue = "VLDB",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadivldb05.pdf",
  abstract = "This paper presents a set of algorithms for efficiently evaluating join queries over static data tables in sensor networks. We describe and evaluate three algorithms that take advantage of distributed join techniques. Our algorithms are capable of running in limited amounts of RAM, can distribute the storage burden over groups of nodes, and are tolerant to dropped packets and node failures. REED is thus suitable for a wide range of event-detection applications that traditional sensor network database and data collection systems cannot be used to implement.",
  pdfKB = "292",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Analytical database systems,Distributed systems,Streaming data,Stream database systems",
}

This paper presents a set of algorithms for efficiently evaluating join queries over static data tables in sensor networks. We describe and evaluate three algorithms that take advantage of distributed join techniques. Our algorithms are capable of running in limited amounts of RAM, can distribute the storage burden over groups of nodes, and are tolerant to dropped packets and node failures. REED is thus suitable for a wide range of event-detection applications that traditional sensor network database and data collection systems cannot be used to implement.

@inproceedings{cstore,
  author    = {Michael Stonebraker and Daniel J. Abadi and Adam Batkin and Xuedong Chen and Mitch Cherniack and Miguel Ferreira and Edmond Lau and Amerson Lin and Samuel R. Madden and Elizabeth J. O'Neil and Patrick E. O'Neil and Alexander Rasin and Nga Tran and Stan B. Zdonik},
  title     = {C-Store: A Column-Oriented DBMS},
  booktitle = {VLDB},
  year      = {2005},
  address   = {Trondheim, Norway},
  pages     = {553-564},
  venue = "VLDB",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldb.pdf",
  abstract = "This paper presents the design of a read-optimized relational DBMS that contrasts sharply with most current systems, which are write-optimized. Among the many differences in its design are: storage of data by column rather than by row, careful coding and packing of objects into storage including main memory during query processing, storing an overlapping collection of column-oriented projections, rather than the current fare of tables and indexes, a non-traditional implementation of transactions which includes high availability and snapshot isolation for read-only transactions, and the extensive use of bitmap indexes to complement B-tree structures. We present preliminary performance data on a subset of TPC-H and show that the system we are building, C-Store, is substantially faster than popular commercial products. Hence, the architecture looks very encouraging.",
  pdfKB = "170",
  publicationtype = "Conference Paper",
  award = "10 Year Best Paper Award",
  note = "<u>10 Year Best Paper Award</u>",
  displayCategory = "Conference or Journal Publication",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
}

This paper presents the design of a read-optimized relational DBMS that contrasts sharply with most current systems, which are write-optimized. Among the many differences in its design are: storage of data by column rather than by row, careful coding and packing of objects into storage including main memory during query processing, storing an overlapping collection of column-oriented projections, rather than the current fare of tables and indexes, a non-traditional implementation of transactions which includes high availability and snapshot isolation for read-only transactions, and the extensive use of bitmap indexes to complement B-tree structures. We present preliminary performance data on a subset of TPC-H and show that the system we are building, C-Store, is substantially faster than popular commercial products. Hence, the architecture looks very encouraging.

@inproceedings{borealis,
  author    = {Daniel J. Abadi and Yanif Ahmad and Magdalena Balazinska and Ugur Cetintemel and Mitch Cherniack and Jeong-Hyon Hwang and Wolfgang Lindner and Anurag S. Maskey and Alexander Rasin and Esther Ryvkina and Nesime Tatbul and Ying Xing and Stan B. Zdonik},
  title     = {The Design of the Borealis Stream Processing Engine},
  booktitle = {CIDR},
  year      = {2005},
  address  = {Asilomar, CA, USA},
  venue = "CIDR",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/cidr05.pdf",
  abstract = "Borealis is a second-generation distributed stream processing engine that is being developed at Brandeis University, Brown University, and MIT. Borealis inherits core stream processing functionality from Aurora and distribution functionality from Medusa. Borealis modifies and extends both systems in non-trivial and critical ways to provide advanced capabilities that are commonly required by newly-emerging stream processing applications. In this paper, we outline the basic design and functionality of Borealis. Through sample real-world applications, we motivate the need for dynamically revising query results and modifying query specifications. We then describe how Borealis addresses these challenges through an innovative set of features, including revision records, time travel, and control lines. Finally, we present a highly flexible and scalable QoS-based optimization model that operates across server and sensor networks and a new fault-tolerance model with flexible consistency-availability trade-offs.",
  pdfKB = "143",
  publicationtype = "Conference Paper",
  displayCategory = "Conference or Journal Publication",
  keywords = "Streaming data, Stream database systems,Distributed systems",
  project:multiple = "Aurora",
}

Borealis is a second-generation distributed stream processing engine that is being developed at Brandeis University, Brown University, and MIT. Borealis inherits core stream processing functionality from Aurora and distribution functionality from Medusa. Borealis modifies and extends both systems in non-trivial and critical ways to provide advanced capabilities that are commonly required by newly-emerging stream processing applications. In this paper, we outline the basic design and functionality of Borealis. Through sample real-world applications, we motivate the need for dynamically revising query results and modifying query specifications. We then describe how Borealis addresses these challenges through an innovative set of features, including revision records, time travel, and control lines. Finally, we present a highly flexible and scalable QoS-based optimization model that operates across server and sensor networks and a new fault-tolerance model with flexible consistency-availability trade-offs.

@misc{aurora,
  author = {Daniel J. Abadi and Don Carney and  Ugur Cetintemel and Mitch Cherniack and Christian Convey and Sangdon Lee and Michael Stonebraker and Nesime Tatbul and Stan B. Zdonik},
  title = {Aurora: A New Model and Architecture for Data Stream Management},
  howpublished = {VLDB Journal, 12(2)},
  month = {September},
  year = {2003},
  pages = {120-139},
  venue = {VLDB Journal},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldb095.pdf",
  abstract = "This paper describes the basic processing model and architecture of Aurora, a new system to manage data streams for monitoring applications. Monitoring applications differ substantially from conventional business data processing. The fact that a software system must process and react to continual inputs from many sources (e.g., sensors) rather than from human operators requires one to rethink the fundamental architecture of a DBMS for this application area. In this paper, we present Aurora, a new DBMS currently under construction at Brandeis University, Brown University, and M.I.T. We first provide an overview of the basic Aurora model and architecture and then describe in detail a stream-oriented set of operators.",
  pdfKB = "984",
  publicationtype = "Journal Article",
  displayCategory = "Conference or Journal Publication",
  keywords = "Streaming data, Stream database systems",
  project:multiple = "Aurora",
}

%% Technical Reports

This paper describes the basic processing model and architecture of Aurora, a new system to manage data streams for monitoring applications. Monitoring applications differ substantially from conventional business data processing. The fact that a software system must process and react to continual inputs from many sources (e.g., sensors) rather than from human operators requires one to rethink the fundamental architecture of a DBMS for this application area. In this paper, we present Aurora, a new DBMS currently under construction at Brandeis University, Brown University, and M.I.T. We first provide an overview of the basic Aurora model and architecture and then describe in detail a stream-oriented set of operators.

@misc{hadoopdb-demo,
        title = {HadoopDB in Action: Building Real World Applications},
        year = {2010},
        howpublished = {Demonstration. SIGMOD},
        venue = {SIGMOD},
        address = {Indianapolis, USA},
        author = {Azza Abouzied and Kamil Bajda-Pawlikowski and Jiewen Huang and Daniel J. Abadi and Avi Silberschatz},
	publicationtype = {Demonstration},
	pdfKB = {520},
	url_Paper = {http://www.cs.umd.edu/~abadi/papers/hadoopdb-demo.pdf},
        displayCategory = "Demonstration",
        iis0845643 = "Supported papers",
        nsfclue = "Supported papers",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce",
        project:multiple = "HadoopDB",
}

@misc{hstore-demo,
        title = {H-Store: A High-Performance, Distributed Main Memory Transaction Processing System},
        year = {2008},
        howpublished = {Demonstration. VLDB},
        venue = {VLDB},
        address = {Aukland, New Zealand},
        author = {Robert Kallman and Hideaki Kimura and Jonathan Natkins and Andrew Pavlo and Alex Rasin and Stan Zdonik and Evan Jones and Yang Zhang and Samuel Madden and Michael Stonebraker and John Hugg and Daniel J. Abadi},
	publicationtype = {Demonstration},
	pdfKB = {440},
	url_Paper = {http://www.cs.umd.edu/~abadi/papers/hstore-demo.pdf},
        displayCategory = "Demonstration",
        keywords = "Scalable transactions,Multi-core systems,Main-memory database systems,Distributed systems,H-Store",
        project:multiple = "H-Store",
}

@misc{sensor-stream-integration,
  author    = {Daniel J. Abadi and Wolfgang Lindner and Samuel R. Madden and Jorg Schuler},
  title     = {An Integration Framework for Sensor Networks and Data Stream Management Systems},
  howpublished = {Demonstration. VLDB},
  year      = {2004},
  pages     = {1361-1364},
  address   = {Toronto, Canada},
  venue = "VLDB",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldb04.pdf",
  abstract = "This demonstration shows an integrated query processing environment where users can seamlessly query both a data stream management system and a sensor network with one query expression. By integrating the two query processing systems, the optimization goals of the sensor network (primarily power) and server network (primarily latency and quality) can be unified into one quality of service metric.",
  pdfKB = "116",
  publicationtype = "Demonstration",
  displayCategory = "Demonstration",
  keywords = "Distributed systems,Streaming data, Stream database systems",
}

This demonstration shows an integrated query processing environment where users can seamlessly query both a data stream management system and a sensor network with one query expression. By integrating the two query processing systems, the optimization goals of the sensor network (primarily power) and server network (primarily latency and quality) can be unified into one quality of service metric.

@misc{aurora-demo,
  author    = {Daniel J. Abadi and Don Carney and Ugur Cetintemel and Mitch Cherniack and Christian Convey and Christina Erwin and Eddie Galvez and Matt Hatoun and Jeong-Hyon Hwang and Anurag S. Maskey and Alexander Rasin and A. Singer and Michael Stonebraker and Nesime Tatbul and Ying Xing and R. Yan and Stan B. Zdonik},
  title     = {Aurora: A Data Stream Management System},
  howpublished = {Demonstration. SIGMOD},
  year      = {2003},
  pages     = {666-666},
  address   = {San Diego, CA, USA},
  venue = "SIGMOD",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/AuroraDemo.pdf",
  abstract = "Streams are continuous data feeds generated by such sources as sensors, satellites, and stock feeds. Monitoring applications track data from numerous streams, filtering them for signs of abnormal activity, and processing them for purposes of filtering, aggregation, reduction, and correlation. Aurora is a general-purpose data stream manager that is being designed and implemented (at Brandeis University, Brown University, and M.I.T.) to efficiently support a variety of real-time monitoring applications.",
  pdfKB = "151",
  publicationtype = "Demonstration",
  displayCategory = "Demonstration",
  keywords = "Streaming data, Stream database systems",
  project:multiple = "Aurora",
}

Streams are continuous data feeds generated by such sources as sensors, satellites, and stock feeds. Monitoring applications track data from numerous streams, filtering them for signs of abnormal activity, and processing them for purposes of filtering, aggregation, reduction, and correlation. Aurora is a general-purpose data stream manager that is being designed and implemented (at Brandeis University, Brown University, and M.I.T.) to efficiently support a variety of real-time monitoring applications.

@misc{vcoko,
  author    = {Daniel J. Abadi and Mitch Cherniack},
  title     = {Visual COKO: A Debugger for Query Optimizer Development},
  howpublished = {Demonstration. SIGMOD},
  year      = {2002},
  address  = {Madison, Wisconsin},
  pages    = {617-617},
  venue = "SIGMOD",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/vcoko.pdf",
  abstract = "Query optimization generates plans to retrieve data requested by queries. Query rewriting, which is the first step of this process, rewrites a query expression into an equivalent form to prepare it for plan generation. COKO-KOLA introduced a new approach to query rewriting that enables query rewrites to be formally verified using an automated theorem prover. KOLA is a language for expressing term rewriting rules that can be 'fired' on query expressions. COKO is a language for expressing query rewriting transformations that are too complex to express with simple KOLA rules. COKO is a programming language designed for query optimizer development. Programming languages require debuggers, and in this demonstration, we illustrate our COKO debugger: Visual COKO. Visual COKO enables a query optimization developer to visually trace the execution of a COKO transformation. At every step of the transformation, the developer can view a tree-display that illustrates how the original query expression has evolved.",
  pdfKB = "111",
  publicationtype = "Demonstration",
  displayCategory = "Demonstration",
  keywords = "Analytical database systems",
}

%% Theses

Query optimization generates plans to retrieve data requested by queries. Query rewriting, which is the first step of this process, rewrites a query expression into an equivalent form to prepare it for plan generation. COKO-KOLA introduced a new approach to query rewriting that enables query rewrites to be formally verified using an automated theorem prover. KOLA is a language for expressing term rewriting rules that can be 'fired' on query expressions. COKO is a language for expressing query rewriting transformations that are too complex to express with simple KOLA rules. COKO is a programming language designed for query optimizer development. Programming languages require debuggers, and in this demonstration, we illustrate our COKO debugger: Visual COKO. Visual COKO enables a query optimization developer to visually trace the execution of a COKO transformation. At every step of the transformation, the developer can view a tree-display that illustrates how the original query expression has evolved.

@misc{abadi-anaphora,
  author    = {Daniel J. Abadi},
  title     = {Comparing Domain-Specific and Non-Domain-Specific Anaphora Resolution Techniques},
  howpublished = {Cambridge University MPhil Dissertation},
  year      = {2003},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/FinalMPhil.pdf",
  abstract = "Three different pronominal anaphora resolution techniques are examined. The first two techniques compare traditional salience-based approaches when different amounts of syntactic information are available. The improvement in pronoun resolution precision is quantified when a large scale grammar is used to extract detailed syntactic information rather than inferring this information robustly using pattern matching. The third technique uses domain knowledge instead of syntactic information to resolve pronouns. The domain knowledge required for this algorithm can be automatically acquired from a database backend schema representation of the domain. Each of these three techniques is evaluated separately, and then the domain-specific and non-domain-specific algorithms are combined and evaluated.",
  pdfKB = "164",
  publicationtype = "Thesis",
  note = "M.Phil. Thesis",
  displayCategory = "M.Phil. Thesis",
}

Three different pronominal anaphora resolution techniques are examined. The first two techniques compare traditional salience-based approaches when different amounts of syntactic information are available. The improvement in pronoun resolution precision is quantified when a large scale grammar is used to extract detailed syntactic information rather than inferring this information robustly using pattern matching. The third technique uses domain knowledge instead of syntactic information to resolve pronouns. The domain knowledge required for this algorithm can be automatically acquired from a database backend schema representation of the domain. Each of these three techniques is evaluated separately, and then the domain-specific and non-domain-specific algorithms are combined and evaluated.

@misc{abadi-phd,
  author = {Daniel J. Abadi},
  title = {Query Execution in Column-Oriented Database Systems},
  howpublished = {MIT PhD Dissertation},
  year = {2008},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/abadiphd.pdf",
  abstract = "There are two obvious ways to map a two-dimension relational database table onto a one-dimensional storage interface: store the table row-by-row, or store the table column-by-column. Historically, database system implementations and research have focused on the row-by row data layout, since it performs best on the most common application for database systems: business transactional data processing. However, there are a set of emerging applications for database systems for which the row-by-row layout performs poorly. These applications are more analytical in nature, whose goal is to read through the data to gain new insight and use it to drive decision making and planning. In this dissertation, we study the problem of poor performance of row-by-row data layout for these emerging applications, and evaluate the column-by-column data layout opportunity as a solution to this problem. There have been a variety of proposals in the literature for how to build a database system on top of column-by-column layout. These proposals have different levels of implementation effort, and have different performance characteristics. If one wanted to build a new database system that utilizes the column-by-column data layout, it is unclear which proposal to follow. This dissertation provides (to the best of our knowledge) the only detailed study of mutliple implementation approaches of such systems, categorizing the different approaches into three broad categories, and evaluating the tradeoffs between approaches. We conclude that building a query executer specifically designed for the column-by-column query layout is essensial to acheive good performance. Consequently, we describe the implementation of C-Store, a new database system with a storage layer and query executer built for column-by-column data layout. We introduce three new query execution technqiues that significantly improve performance. First, we look at the problem of integrating compression and execution so that the query executer is capable of directly operating on compressed data. This improves performance by improving I/O (less data needs to be read off disk), and CPU (the data need not be decompressed). We describe our solution to the problem of executer extensibility - how can new compression techniques be added to the system without having to rewrite the operator code? Second, we analyze the problem of tuple construction (stitching together attributes from multiple columns into a row-oriented ``tuple''). Tuple construction is required when operators need to access multiple attributes from the same tuple; however, if done at the wrong point in a query plan, a significant performance penalty is paid. We introduce an analytical model and some heuristics to use that help decide when in a query plan tuple construction should occur. Third, we introduce a new join technique, the ``invisible join'' that improves performance of a specific type of join that is common in the applications for which column-by-column data layout is a good idea. Finally, we benchmark performance of the complete C-Store database system against other column-oriented database system implementation approachs, and against row-oriented databases. We benchmark two applications. The first application is a typical analytical application for which column-by-column data layout is known to outperform row-by-row data layout. The second application is another emerging application, the Semantic Web, for which column-oriented database systems are not currently used. We find that on the first application, the complete C-Store system performed 10 to 18 times faster than alternative column-store implementation approaches, and 6 to 12 times faster than a commercial database system that uses a row-by-row data layout. On the Semantic Web application, we find that C-Store outperforms other state-of-the-art data management techniques by an order of magnitude, and outperforms other common data management technqiues by almost two orders of magnitude. Benchmark queries, which used to take multiple minutes to execute, can now be answered in several seconds.",
  pdfKB = "1301",
  publicationtype = "Thesis",
  oldnote = "PhD Thesis",
  award = "SIGMOD Jim Gray Doctoral Dissertation Award (2008)",
  note = "<u>SIGMOD Jim Gray Doctoral Dissertation Award</u>",
  displayCategory = "Phd Thesis",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store",
  project:multiple = "C-Store",
}

There are two obvious ways to map a two-dimension relational database table onto a one-dimensional storage interface: store the table row-by-row, or store the table column-by-column. Historically, database system implementations and research have focused on the row-by row data layout, since it performs best on the most common application for database systems: business transactional data processing. However, there are a set of emerging applications for database systems for which the row-by-row layout performs poorly. These applications are more analytical in nature, whose goal is to read through the data to gain new insight and use it to drive decision making and planning. In this dissertation, we study the problem of poor performance of row-by-row data layout for these emerging applications, and evaluate the column-by-column data layout opportunity as a solution to this problem. There have been a variety of proposals in the literature for how to build a database system on top of column-by-column layout. These proposals have different levels of implementation effort, and have different performance characteristics. If one wanted to build a new database system that utilizes the column-by-column data layout, it is unclear which proposal to follow. This dissertation provides (to the best of our knowledge) the only detailed study of mutliple implementation approaches of such systems, categorizing the different approaches into three broad categories, and evaluating the tradeoffs between approaches. We conclude that building a query executer specifically designed for the column-by-column query layout is essensial to acheive good performance. Consequently, we describe the implementation of C-Store, a new database system with a storage layer and query executer built for column-by-column data layout. We introduce three new query execution technqiues that significantly improve performance. First, we look at the problem of integrating compression and execution so that the query executer is capable of directly operating on compressed data. This improves performance by improving I/O (less data needs to be read off disk), and CPU (the data need not be decompressed). We describe our solution to the problem of executer extensibility - how can new compression techniques be added to the system without having to rewrite the operator code? Second, we analyze the problem of tuple construction (stitching together attributes from multiple columns into a row-oriented ``tuple''). Tuple construction is required when operators need to access multiple attributes from the same tuple; however, if done at the wrong point in a query plan, a significant performance penalty is paid. We introduce an analytical model and some heuristics to use that help decide when in a query plan tuple construction should occur. Third, we introduce a new join technique, the ``invisible join'' that improves performance of a specific type of join that is common in the applications for which column-by-column data layout is a good idea. Finally, we benchmark performance of the complete C-Store database system against other column-oriented database system implementation approachs, and against row-oriented databases. We benchmark two applications. The first application is a typical analytical application for which column-by-column data layout is known to outperform row-by-row data layout. The second application is another emerging application, the Semantic Web, for which column-oriented database systems are not currently used. We find that on the first application, the complete C-Store system performed 10 to 18 times faster than alternative column-store implementation approaches, and 6 to 12 times faster than a commercial database system that uses a row-by-row data layout. On the Semantic Web application, we find that C-Store outperforms other state-of-the-art data management techniques by an order of magnitude, and outperforms other common data management technqiues by almost two orders of magnitude. Benchmark queries, which used to take multiple minutes to execute, can now be answered in several seconds.

@techreport{barton-benchmark,
  title = {Using The Barton Libraries Dataset As An RDF Benchmark},
  number = {MIT-CSAIL-TR-2007-036},
  institution = {MIT},
  year      = {2007},
  author = {Daniel J. Abadi and Adam Marcus and Samuel R. Madden and Kate Hollenbach},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/bench.pdf",
  abstract = "This report describes the Barton Libraries RDF dataset and Longwell query benchmark that we use for our recent VLDB paper on Scalable Semantic Web Data Management Using Vertical Partitioning",
  pdfKB = "357",
  publicationtype = "Technical Report",
  displayCategory = "Technical Report",
  keywords = "Scalable queries,Parallel database systems,Analytical database systems,Graph database systems,Non-relational data,C-Store",
  project:multiple = "C-Store",
  datasetURL = "http://dslam.cs.umd.edu/data/barton",
}

%% Tutorials and Panels

This report describes the Barton Libraries RDF dataset and Longwell query benchmark that we use for our recent VLDB paper on Scalable Semantic Web Data Management Using Vertical Partitioning

@inproceedings{columnstore-tutorial,
        title = "Tutorial: Column oriented Database Systems",
        year = "2009",
        booktitle = "VLDB",
        venue = "VLDB",
        author = {Daniel J. Abadi and Peter A. Boncz and Stavros Harizopoulos},
	publicationtype = "Tutorial",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/columnstore-tutorial.pdf",
	pdfKB = "23",
	url_Slides = "http://www.slideshare.net/abadid/vldb-2009-tutorial-on-columnstores",
        displayCategory = "Tutorial or Panel",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,Column-stores,C-Store",
        project:multiple = "C-Store",
}

@inproceedings{sql-on-hadoop-tutorial,
        title = "Tutorial: SQL-on-Hadoop Systems",
        year = "2015",
        booktitle = "VLDB",
        venue = "VLDB",
        author = {Daniel J. Abadi and Shivnath Babu and Fatma Ozcan and Ippokratis Pandis},
	publicationtype = "Tutorial",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/sql-on-hadoop-tutorial.pdf",
        url_Slides = "https://www.slideshare.net/abadid/sqlonhadoop-tutorial",
	pdfKB = "144",
        displayCategory = "Tutorial or Panel",
        keywords = "Scalable queries,Parallel database systems,Analytical database systems,HadoopDB,MapReduce",
        project:multiple = "HadoopDB",
}

@inproceedings{bigdata-bubble-panel,
  author    = {Fatma Ozcan and
               Nesime Tatbul and
               Daniel J. Abadi and
               Marcel Kornacker and
               C. Mohan and
               Karthik Ramasamy and
               Janet L. Wiener},
  title     = {Panel: Are we experiencing a big data bubble?},
  booktitle = {SIGMOD},
  venue = {SIGMOD},
  pages     = {1407-1408},
  year      = {2014},
  publicationtype = "Panel Presentation",
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/sigmod2014-panel.pdf",
  pdfKB = "625",
  displayCategory = "Tutorial or Panel",
}

@inproceedings{abadi-vldb09-panel,
        title = "Panel: How best to build web-scale data managers?",
        year = "2009",
        booktitle = "VLDB",
        venue = "VLDB",
        author = {Philip A. Bernstein and Daniel J. Abadi and Michael J. Cafarella and Joseph M. Hellerstein and Donald Kossmann and Samuel Madden},
	publicationtype = "Panel Presentation",
	url_Paper = "http://www.cs.umd.edu/~abadi/papers/vldb2009-panel.pdf",
	pdfKB = "600",
	url_Slides = "http://www.slideshare.net/abadid/daniel-abadi-vldb-2009-panel",
        displayCategory = "Tutorial or Panel",

}

%% Blog Posts

@misc{abadi-ugrad-thesis,
  author    = {Daniel J. Abadi},
  title     = {Visual COKO: A Debugger for Query Optimizer Development},
  howpublished = {Brandeis University Senior Honors Thesis},
  year      = {2002},
  url_Paper = "http://www.cs.umd.edu/~abadi/papers/VisualCOKOThesis.pdf",
  abstract = "Query optimization generates plans to retrieve data requested by queries, and query rewriting (rewriting a query expression into an equivalent form to prepare it for plan generation) is typically the first step. COKO-KOLA introduced a new approach to query rewriting that enables query rewrites to be formally verified using an automated theorem prover. KOLA is a language for expressing rewriting rules that can be fired on query expressions. COKO is a language for expressing query rewriting transformations that are too complex to express with simple KOLA rules. COKO is a programming language designed for query optimizer development. Programming languages require debuggers, and this paper describes a COKO debugger: Visual COKO. Visual COKO enables a query optimization developer to visually trace the execution of a COKO transformation. At every step of the transformation, the developer can view a tree-display that illustrates how the original query expression has evolved. Rule-based query rewriting and the COKO-KOLA project are described for background. Then the COKO syntax is summarized from the point of view of the COKO programmer using an example query transformation that converts query predicates to conjunctive normal form. Visual COKO is described and instructions for its use are presented. Finally, a description of its implementation is given.",
  pdfKB = "132",
  publicationtype = "Thesis",
  note = "Undergraduate Thesis",
  displayCategory = "Undergraduate Senior Honors Thesis",
  keywords = "Analytical database systems",
}

Query optimization generates plans to retrieve data requested by queries, and query rewriting (rewriting a query expression into an equivalent form to prepare it for plan generation) is typically the first step. COKO-KOLA introduced a new approach to query rewriting that enables query rewrites to be formally verified using an automated theorem prover. KOLA is a language for expressing rewriting rules that can be fired on query expressions. COKO is a language for expressing query rewriting transformations that are too complex to express with simple KOLA rules. COKO is a programming language designed for query optimizer development. Programming languages require debuggers, and this paper describes a COKO debugger: Visual COKO. Visual COKO enables a query optimization developer to visually trace the execution of a COKO transformation. At every step of the transformation, the developer can view a tree-display that illustrates how the original query expression has evolved. Rule-based query rewriting and the COKO-KOLA project are described for background. Then the COKO syntax is summarized from the point of view of the COKO programmer using an example query transformation that converts query predicates to conjunctive normal form. Visual COKO is described and instructions for its use are presented. Finally, a description of its implementation is given.

@misc{dbreport-2022,
  author    = {Daniel Abadi and
               Anastasia Ailamaki and
               David G. Andersen and
               Peter Bailis and
               Magdalena Balazinska and
               Philip A. Bernstein and
               Peter A. Boncz and
               Surajit Chaudhuri and
               Alvin Cheung and
               AnHai Doan and
               Luna Dong and
               Michael J. Franklin and
               Juliana Freire and
               Alon Y. Halevy and
               Joseph M. Hellerstein and
               Stratos Idreos and
               Donald Kossmann and
               Tim Kraska and
               Sailesh Krishnamurthy and
               Volker Markl and
               Sergey Melnik and
               Tova Milo and
               C. Mohan and
               Thomas Neumann and
               Beng Chin Ooi and
               Fatma Ozcan and
               Jignesh M. Patel and
               Andrew Pavlo and
               Raluca A. Popa and
               Raghu Ramakrishnan and
               Christopher R{\'{e}} and
               Michael Stonebraker and
               Dan Suciu},
  title     = {The Seattle report on database research},
  journal   = {Commun. {ACM}},
  volume    = {65},
  number    = {8},
  pages     = {72--79},
  year      = {2022},
  doi       = {10.1145/3524284},
  howpublished = {CACM 65(8)},
  venue = "CACM",
}