#### DMCA

## 1ILP Modulo Data

Citations: | 1 - 0 self |

### Citations

1274 | A relational model of data for large shared data banks,
- Codd
- 1970
(Show Context)
Citation Context ...ch, and discuss potential applications. I. INTRODUCTION In 2010, enterprises and users stored more than 13 exabytes of new data [1]. Database Management Systems (DBMS’s) based on the Relational Model =-=[3]-=- are a key component in the computing infrastructure of virtually any organization. With big data playing a determining role in business and science, we are motivated to rethink data management and an... |

586 |
The complexity of relational query languages, in
- Vardi
- 1982
(Show Context)
Citation Context ...t within the scope of this paper, and has mostly theoretical significance. In practice, query size is orders of magnitude smaller than data size. Conversely, it is meaningful to study data complexity =-=[19]-=-, i.e., complexity where only the amount of data varies. Instead of assuming a query of constant size, we provide a stronger result by limiting the number of tables that can participate in a cross pro... |

455 | Simplification by cooperating decision procedures,
- Nelson, Oppen
- 1979
(Show Context)
Citation Context ...we systematically explore the space of (dis)equalities between these variables. This exchange of information resembles the non-deterministic Nelson-Oppen scheme (NO) for combining decision procedures =-=[15]-=-. We demonstrate that NO can accommodate membership constraints. Definition 5 (Arrangement). Let E be an equivalence relation over a set of variables V . The set α(V,E) = {x = y | xEy} ∪ {x 6= y | x, ... |

255 | Solving SAT and SAT Modulo Theories: From an Abstract DavisPutnam-Logemann-Loveland Procedure to DPLL(T)",
- Nieuwenhuis, Oliveras, et al.
- 2006
(Show Context)
Citation Context ...ementing such a combination. BC(T ) can easily accommodate data-aware propagation (Section V-A) and branching (Section V-B). Our techniques would be harder to implement within a DPLL(T )-style solver =-=[16]-=-, given that the toplevel search of DPLL(T ) is over the Booleans (and not the integers). A DPLL(T )-based implementation of our techniques would essentially require integrating branch-and-bound in DP... |

101 | Constraint Integer Programming.
- Achterberg
- 2007
(Show Context)
Citation Context ...in linear time. It is thus vital that our techniques yield acceptable performance. Equations 7 and 8 bound x to [min{1, 2, 3, 4},max{1, 2, 3, 4}] = [1, 4] and y to [min{2, 4, 6, 8},max{2, 4, 6, 8}] = =-=[2, 8]-=-. Given the equality x = y, ILP propagation deduces that x, y ∈ [2, 4], since [2, 4] is the intersection of permissible ranges for x and y. The membership procedure detects that match now only holds f... |

31 | Data Structures for Generalised Arc Consistency for Extensional Constraints. In
- Gent, Jefferson, et al.
- 2007
(Show Context)
Citation Context ...ery over concrete tuples that any DBMS can evaluate in linear time. It is thus vital that our techniques yield acceptable performance. Equations 7 and 8 bound x to [min{1, 2, 3, 4},max{1, 2, 3, 4}] = =-=[1, 4]-=- and y to [min{2, 4, 6, 8},max{2, 4, 6, 8}] = [2, 8]. Given the equality x = y, ILP propagation deduces that x, y ∈ [2, 4], since [2, 4] is the intersection of permissible ranges for x and y. The memb... |

16 | Tiresias: The database oracle for how-to queries.
- Meliou, Suciu
- 2012
(Show Context)
Citation Context ...d function of multiple quantities including a base salary, benefits based on age, employee level (junior, middle, or senior), and bonuses. The management consults the database administrator on how to =-=[13]-=- ameliorate the seeming injustice. Together, they explore bonuses that would allow young employees to exceed the $60000 limit. This amounts to synthesizing tuples for the table of bonuses. An alternat... |

15 | Combining Decision Procedures.
- Manna, Zarba
- 2002
(Show Context)
Citation Context ... needed, since ∃ always appears with positive polarity (as per Definition 2). Membership constraints do not contain arbitrary arithmetic expressions, but only variable symbols. “Variable abstraction” =-=[9]-=- eliminates richer expressions. While variable abstraction allows for compositional reasoning and helps with theoretical analysis, a limited fragment of arithmetic in membership constraints yields mor... |

3 |
Bassem Elkarablieh, Yai Laleye, and Sarfraz Khurshid. Query-Aware Test Generation Using a Relational Constraint Solver
- Khalek
- 2008
(Show Context)
Citation Context ...et applications involve bigger tables. Other approaches tackle constraints arising in database applications with off-the-shelf generic solvers (via eager reductions). Notably, Khalek et al. use Alloy =-=[7]-=-, while Meliou and Suciou use MILP [13]. In neither of these approaches does the core of the solver exploit the structure of database instances, e.g., for branching or propagation. VIII. CONCLUSIONS A... |

3 |
Lecoutre and Radoslaw Szymanek. Generalized arc consistency for positive table constraints
- Christophe
- 2006
(Show Context)
Citation Context ...in linear time. It is thus vital that our techniques yield acceptable performance. Equations 7 and 8 bound x to [min{1, 2, 3, 4},max{1, 2, 3, 4}] = [1, 4] and y to [min{2, 4, 6, 8},max{2, 4, 6, 8}] = =-=[2, 8]-=-. Given the equality x = y, ILP propagation deduces that x, y ∈ [2, 4], since [2, 4] is the intersection of permissible ranges for x and y. The membership procedure detects that match now only holds f... |

2 |
Opportunities with Big Data, 2012. Computing Community Consortium White Paper
- Challenges
(Show Context)
Citation Context ...s use of database techniques. We also experimentally evaluate our approach, and discuss potential applications. I. INTRODUCTION In 2010, enterprises and users stored more than 13 exabytes of new data =-=[1]-=-. Database Management Systems (DBMS’s) based on the Relational Model [3] are a key component in the computing infrastructure of virtually any organization. With big data playing a determining role in ... |

2 |
Data Mining the SDSS SkyServer Database. arXiv preprint cs/0202014,
- Gray, Szalay, et al.
- 2002
(Show Context)
Citation Context ...l Programming tools with the ability to directly access data, thus assisting data-backed decision making. Such tools would also benefit scientists in fields ranging from ornithology [17] to astronomy =-=[5]-=-, by providing immediate feedback on the consistency between models the scientists devise and datasets of observations they collect. Our paper outlines potential applications, while our experimental e... |

1 |
Moura and Nikolaj Bjorner. Efficient E-matching for SMT solvers
- De
- 2007
(Show Context)
Citation Context ...t al. describe the Qex technique and tool that uses Z3 to generate tests for SQL queries [20]. Qex essentially encodes the relational operators via axioms, which are later instantiated via E-matching =-=[14]-=-. E-matching is a generic scheme that is not optimized in any way for database problems. Qex is geared towards relatively small tables that suffice as test cases, while our target applications involve... |