Fifth International Workshop on
Testing Database Systems

Co-located with ACM SIGMOD 2012


  • PODS Keynote


  • Coffee Break


  • Welcome
  • Lightning Round

10:45AM  ---  Session 1: Dealing with Big Data

  • BloomUnit: Declarative Testing for Distributed Programs
    • Peter Alvaro, UC Berkeley; Andrew Hutchinson, UC Berkeley; Neil Conway, UC Berkeley; William Marczak, UC Berkeley; Joseph Hellerstein, UC Berkeley
  • Automatic Capture of Minimal, Portable, and Executable Bug Repros Using AMPERE [PPT]
    • Lyublena Antova, EMC; Konstantinos Krikellas, EMC; Florian Waas, EMC
  • SCOPE Playback: Self-Validation in the Cloud
    • Ming-Chuan Wu, Microsoft Corporation; Jingren Zhou, Microsoft Corporation; Nicolas Bruno, Microsoft Corporation; Grace Zhang, Microsoft Corporation; Jon Fowler, Microsoft Corporation


  • Lunch

1:00PM  ---  Session 2: Test Artifact Generation

  • CODD: COnstructing Dataless Databases
    • Rakshit Trivedi, Indian Institute of Science, India; Nilavalagan I, Indian Institute of Science, India; Jayant Haritsa, Indian Institute of Science, India 
  • Targeted Genetic Test SQL Generation for the DB2 Database 
    • Dominic Letarte, É. Polytechnique de Montréal; François Gauthier, É. Polytechnique de Montréal; Ettore Merlo, É. Polytechnique de Montréal; Nattavut Sutyanyong, IBM Canada Ltd.; Calisto Zuzarte, IBM Canada Ltd.
  • Test Input Generation for Database Programs using Relational Constraints
    • Michael Marcozzi, FUNDP; Wim Vanhoof, FUNDP; Jean-Luc Hainaut, FUNDP
  • Combined XML/XQuery generator [PPT]
    • Milos Todic, Microsoft; Branislav Uzelac, Microsoft


  • Coffee Break

2:45PM  ---  Keynote by Ooge de Moor

  • Business Intelligence for Software Quality
    • What makes a great piece of code? Who writes that great code? Are there areas where they could do better, say in using the concurrency API? Objective answers require data analysis; the available data include the source code itself, issue tickets, version history, test results and so forth.

      Semmle is a business intelligence platform for doing such data analysis. I’ll show three different views of the analysis results: one to highlight the quality of individual contributions, another to zoom in on areas of the code that are at risk of failure, and finally an interface for power users to create custom analyses. Each view will be illustrated with a high-profile open source project.

      Underneath the Semmle platform is a modern implementation of object-oriented Datalog that executes complex queries on heterogeneous data at blazing speed.
  • Oege de Moor is the CEO of Semmle. He is also a professor at the University of Oxford, where he leads the Programming Languages research theme.


  • Coffee Break

4:15PM  ---   Session 3: Auto-Tunning

  • Self-diagnosing and self-healing indexes
    • Goetz Graefe, Hewlett-Packard Laboratories; Harumi Kuno, Hewlett-Packard Laboratories; Bernhard Seeger, Philipps-Universität Marburg
  • Automated physical designers: What you see is (not) what you get [PPT]
    • Renata Borovica, EPFL; Ioannis Alagiannis, EPFL; Anastasia Ailamaki, EPFL
  • Fighting back: using observability tools to improve the DBMS (not just diagnose it) [PPT]
    • Ryan Johnson, University of Toronto


  • Break

5:30PM  ---  Session 4: Testing

  • Testing the Accuracy of Query Optimizers
    • Zhongxian Gu, University of California, Davi; Mohamed Soliman, Greenplum/EMC; Florian Waas, EMC [PPT]
  • Testing Cardinality Estimation Models in SQL Server
    • Campbell Fraser, Microsoft; Leo Giakoumakis, Microsoft; Vikas Hamine, Microsoft Corp.; Katherine Moore, Microsoft

Platinum Sponsor

Gold Sponsor

Microsoft SQL Server