Hide metadata

dc.date.accessioned2013-03-12T08:13:02Z
dc.date.available2013-03-12T08:13:02Z
dc.date.issued2005en_US
dc.date.submitted2005-03-10en_US
dc.identifier.citationHaavet, Tomas Are. Rindex. Masteroppgave, University of Oslo, 2005en_US
dc.identifier.urihttp://hdl.handle.net/10852/9241
dc.description.abstractRindex (a short for RAM Index) is designed as a platform to be used top of a traditional database management system (DBMS). The main focus for this project has been on how to maximize the usage of the available main memory by keeping the indexes entirely in RAM. This way we avoid disk I/O when using an index, and therefore increase the efficiency drastically. In Rindex the relations still resides on disk, only the indexes are kept entirely in memory. This distinguishes Rindex from already existing MMDBMS (main memory DBMS). We have designed and implemented a first version of this platform. This version of Rindex is limited in that it only supports a subset of the SQL language, and just utilizes static indexes (i.e. modifying relations or schemas is currently not supported). Rindex contains all the standard components: scanner, parser, optimizer, plan generator, plan selector, and plan executor. On startup, the platform generates an index on every attribute that are comparable, e.g. integer, floats, and strings. Inverted indexes are also generated. The main focus of this thesis is to analyzes the efficiency of the substitution operator, a special operator in Rindex. The operator substitutes the attribute values in a intermediate index, with the corresponding values from another attribute (i.e. values with the same primary key). We propose several different solutions to optimize this operator, and this results in a significantly improved substitution operator. We design new data structures for the inverted index, since the substitution algorithm heavily depends and follows from this structure. We also analyzes the algorithm that inserts the substitution operator into the query tree. Our attempt to improve the algorithm, i.e. to minimize the number of substitution operators, led to the conclusion that it requires knowledge of the cost estimates and heuristics used by the optimization algorithm. The reason is that the optimization algorithm and this algorithm often disagrees on the ordering of the operators. The thesis is written in Norwegian.nor
dc.language.isonoben_US
dc.titleRindex : RAM-basert indekshåndtering - en effektivitetsstudieen_US
dc.typeMaster thesisen_US
dc.date.updated2008-06-20en_US
dc.creator.authorHaavet, Tomas Areen_US
dc.subject.nsiVDP::420en_US
dc.identifier.bibliographiccitationinfo:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft.au=Haavet, Tomas Are&rft.title=Rindex&rft.inst=University of Oslo&rft.date=2005&rft.degree=Masteroppgaveen_US
dc.identifier.urnURN:NBN:no-10333en_US
dc.type.documentMasteroppgaveen_US
dc.identifier.duo25142en_US
dc.contributor.supervisorRagnar Normannen_US
dc.identifier.bibsys050557777en_US
dc.identifier.fulltextFulltext https://www.duo.uio.no/bitstream/handle/10852/9241/1/thesis.pdf


Files in this item

Appears in the following Collection

Hide metadata