## How to optimize Rscript comprehensions? (2006)

Citations: | 1 - 0 self |

### BibTeX

@MISC{Bredenoord06howto,

author = {Menno Bredenoord and Prof Dr and Paul Klint and Internship Supervisor and Hogeschool Van Amsterdam},

title = {How to optimize Rscript comprehensions?},

year = {2006}

}

### OpenURL

### Abstract

### Citations

8581 |
Introduction to Algorithms
- Cormen, Leiserson, et al.
- 2001
(Show Context)
Citation Context ...eral data structures have been found in the literature which can be used to represent a set as used in Rscript and to optimize the current implementation. We have identified the following: Hash Tables=-=[15]-=-, Binary Search Tree[15], Red-Black Tree[15] , Binomial Heap[15] and Judy Arrays[16,17,18]. Each of these techniques will be investigated and their performance will be discussed. 2.1.2 Evaluation of o... |

1067 |
Principles of database and knowledge-base systems, volume ii
- Ullman
- 1989
(Show Context)
Citation Context ...ctions that are not possible in Rscript. The following techniques are discussed: Commuting Selections, Semantic Query Caching and Peephole Optimization. 5.1. Commuting Selections Commuting Selections =-=[6]-=- is based on performing selection as early as possible. Hereby the intermediate result is smaller and the operation between the relations is calculated in less time. It can be applied to several opera... |

184 | Tan,M.: Semantic data caching and replacement
- Dar, Franklin, et al.
- 1996
(Show Context)
Citation Context ...ed in order to optimize Rscript comprehensions. Several interesting relational database optimization techniques are found and will be discussed, such as Commuting Selections[6], Semantic Query Caching=-=[8,10]-=- and Peephole optimization[12]. This will also be done in the first iteration. The optimization power of these techniques is based on different methods. Some techniques are replacing expensive operato... |

148 | System/R: a relational approach to database management
- Astrahan, al
- 1976
(Show Context)
Citation Context ... is also worth some further investigation. This will be done in the first iteration (see §3.1.1). Other algebraic optimizations found are Product Elimination[1], Filter Hiding[1,5], Evaluating Options=-=[1,4]-=- and Index Introduction[1,2]. Each of these techniques will be studied and documented in detail in this thesis. 2) Relational Database optimization It is well known that comprehensions show a great si... |

59 | Comprehensions, a query notation for DBPLs
- TRINDER
- 1991
(Show Context)
Citation Context ...points are discussed: transformation rules, performance gain and suitability for Rscript. Several techniques which are found in the literature (as mentioned in the background chapter), like: memoising=-=[2]-=- and Evaluating Options[1,4] are not discussed here. This is because their power for optimization is based only on constructions that are not possible in Rscript. The following techniques are discusse... |

32 | Improving list comprehension database queries
- TRINDER, WADLER
- 1989
(Show Context)
Citation Context ...imization techniques that are specifically focused on comprehensions, several algebraic transformations are identified which state that they perform an optimization. For example: Qualifier Interchange=-=[1,2,5]-=- enables the possibility to switch qualifiers within a comprehension. By placing less expensive qualifiers as foremost as possible in the comprehension, the more expensive qualifiers will need to be e... |

32 |
Peephole optimization
- McKeeman
- 1965
(Show Context)
Citation Context ...comprehensions. Several interesting relational database optimization techniques are found and will be discussed, such as Commuting Selections[6], Semantic Query Caching[8,10] and Peephole optimization=-=[12]-=-. This will also be done in the first iteration. The optimization power of these techniques is based on different methods. Some techniques are replacing expensive operators with less-expensive operato... |

24 | Experience with the setl optimizer
- Freudenberger, Schwartz, et al.
- 1983
(Show Context)
Citation Context ...oints are discussed: transformation rules, performance gain and suitability for Rscript. Several techniques found in the literature (as mentioned in the background chapter), such as: Copy Optimization=-=[13]-=- and Code Motion[13] are not discussed here since their power for optimization is based only on constructions that are not possible in Rscript. The following techniques are discussed: Commuting Select... |

15 | Semantic query caching for heterogeneous databases
- Godfrey, Gryz
- 1997
(Show Context)
Citation Context ...ed in order to optimize Rscript comprehensions. Several interesting relational database optimization techniques are found and will be discussed, such as Commuting Selections[6], Semantic Query Caching=-=[8,10]-=- and Peephole optimization[12]. This will also be done in the first iteration. The optimization power of these techniques is based on different methods. Some techniques are replacing expensive operato... |

11 | Algebraic query optimisation for database programming languages
- Poulovassilis, Small
- 1996
(Show Context)
Citation Context ...imization techniques that are specifically focused on comprehensions, several algebraic transformations are identified which state that they perform an optimization. For example: Qualifier Interchange=-=[1,2,5]-=- enables the possibility to switch qualifiers within a comprehension. By placing less expensive qualifiers as foremost as possible in the comprehension, the more expensive qualifiers will need to be e... |

8 | Intensional query optimization - Godfrey, Gryz - 1996 |

4 |
Java Software Structures: Designing and Using Data Structures, 2nd edition
- Lewis, Case
- 2005
(Show Context)
Citation Context ... emphasis will be on data structures who perform these operations efficiently, such as: Hash Tables, Binary Search Tree, Red-Black Trees, Binomial Heap, and Judy Arrays. 7.1.1 Hash Tables With hashing=-=[14,15]-=- elements are stored inside a hash table at a certain position. The location of an element inside this table is determined by a hashing function which delivers an almost unique key. With this key, the... |

2 |
A Tutorial Introduction to RScript — a Relational Approach to Software Analysis (draft
- Klint
- 2005
(Show Context)
Citation Context .... P n and the corresponding comprehension: Comprehension: { v 1 ,..., v n | v 1 ,..., v n : R , P 1 ,..., P n } A more thorough introduction into comprehensions and Rscript in general can be found in =-=[3]-=-. 2.3. Assumptions When researching algebraic optimization techniques, several assumptions have to be made about the environment in which they will be evaluated. This is necessary in order for some te... |

1 |
Common Subexpression Elimination Processing in Multiple-Query Processing
- Chen, Dunham
- 1998
(Show Context)
Citation Context ...he performance. This approach appears to be a promising technique, which will certainly need some more detailed investigation. Another algebraic optimization is called Common Subexpression Elimination=-=[1,11]-=- and states that expressions which are used multiple times, should be calculated once and subsequently the result should be reused. In larger applications the possibility of expressions being used mul... |

1 |
Judy Arrays web page
- Doug, Hewlett-Packard
(Show Context)
Citation Context ... is also shown in figure 1. Merging BHs with different depths can also be done efficient, in O(log n). The main disadvantage of BHs is that the search of an element takes O(n). 7.1.5. Judy Arrays Judy=-=[16]-=- has been brought to our attention with the question whether it can be used as a data structure for Rscript. Judy is a library for C that can be used to replace different types of data structures. The... |

1 |
A Performance Comparison of Judy to Hash Tables
- Sean
- 2003
(Show Context)
Citation Context ...t a set as used in Rscript and to optimize the current implementation. We have identified the following: Hash Tables[15], Binary Search Tree[15], Red-Black Tree[15] , Binomial Heap[15] and Judy Arrays=-=[16,17,18]-=-. Each of these techniques will be investigated and their performance will be discussed. 2.1.2 Evaluation of optimizations The following information about measurements, performance gain and comparison... |

1 |
IV Shop Manual. An internal technical description of Judy. Hewlett-Packard
- Judy
- 2002
(Show Context)
Citation Context ...t a set as used in Rscript and to optimize the current implementation. We have identified the following: Hash Tables[15], Binary Search Tree[15], Red-Black Tree[15] , Binomial Heap[15] and Judy Arrays=-=[16,17,18]-=-. Each of these techniques will be investigated and their performance will be discussed. 2.1.2 Evaluation of optimizations The following information about measurements, performance gain and comparison... |