| Till Westmann, Donald Kossmann, Sven Helmer, Guido Moerkotte. The Implementation and Performance of Compressed Databases. SIGMOD Record 29(3), 2000. |
....block addressing that coalesces multiple list elements, do not apply because the adjacency list of URL ids must be exact. Other techniques, like delta codes, we borrow. Similarly, papers on general database compression focus on reducing disk space and I O requirements of disk based databases [GRS98, WKHM00], not on memory requirements While search engines such as Google [Google] must use compression techniques to store the Web graph, they are unpublished. Altavista [AV] has used our Link2 in production since 1999. Two other groups have published research on compressing the Web graph. Adler and ....
T. Westmann, D. Kossman, S. Helmer, G. Moerkotte. The implementation and performance of compressed databases. Sigmod Record, 29 (3), Sept., 2000, pp 55-67.
....commands. A typical command is ARITH ADD A UI4 ZZZ which adds two unsigned four byte integers found in Z registers and puts the result into another Z register. In general, a command name starts whith a group name followed by the command. Then an optional result mode (borrowed from AVM, see [42]) and a type follow. Last in the command name is a speci fication of the register sets for the arguments and the result. Altogether there are more than 1500 commands that can be interpreted by the NVM. Let us consider a small example of a program that adds two numbers given in X registers 1 and ....
T. Westmann, D. Kossmann, S. Helmer, and G. Moerkotte. The implementation and performance of compressed databases. SIGMOD Record, 29(3):55--67, 2000.
....as well; however, so far we were not able to implement these techniques as part of our work since we relied on the capabilities of standard, off the shelf RDBMSs. Bitmap indexing techniques and database compression techniques are currently being integrated into several relational database systems [11, 10, 14]; therefore, it should be possible to explore the usefulness of these features in future work. 4 Extended XML QL Query Processing We will now turn to a discussion of how XML QL queries with contains predicates can be processed. We will first describe query processing in the first scenario of the ....
T. Westmann, D. Kossmann, S. Helmer, and G. Moerkotte. The implementation and performance of compressed databases. November 1998. Technical Report. University of Mannheim.
....in the query set. So the referenced set cannot be a subset of the query set. 10 We use several well known techniques to increase the performance of inverted lists. To reduce the size of the lists we compress them. We keep the lists sorted and encode the gaps using very light weight techniques [25]. It is also sensible to fetch the lists in increasing size and to use thresholding, i.e. instead of fetching very large lists, we immediately access the data items. For a more detailed explanation of these techniques see [31, 32] Implementation details The root object of an inverted file index ....
T. Westmann, D. Kossmann, S. Helmer, and G. Moerkotte. The implementation and performance of compressed databases. Technical Report 3/98, Universitat Mannheim, 1998. http://pi3.informatik.uni-mannheim.de.
No context found.
Till Westmann, Donald Kossmann, Sven Helmer, Guido Moerkotte. The Implementation and Performance of Compressed Databases. SIGMOD Record 29(3), 2000.
No context found.
T. Westmann, et al., "Implementation and Performance of Compressed Databases", SIGMOD Record, 2000.
No context found.
Till Westmann, Donald Kossmann, Sven Helmer, and Guido Moerkotte. The Implementation and Performance of Compressed Databases. SIGMOD Record, 29(3):55--67, 2000.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC