Fiche du document
23 mai 2014
- handle: 10670/1.lv2xqy
- hal: hal-01350933
- doi: 10.1007/978-3-642-55355-4_28
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-642-55355-4_28
http://creativecommons.org/licenses/by/ , info:eu-repo/semantics/OpenAccess
Citer ce document
Suqing Zhang et al., « Pipeline and Data Parallel Hybrid Indexing Algorithm for Multi-core Platform », HAL-SHS : sciences de l'information, de la communication et des bibliothèques, ID : 10.1007/978-3-642-55355-4_28
Métriques
Partage / Export
Résumé
The scale and growth rate of today’s text collection bring new challenges for index construction. To tackle this problem, Pipeline and Data Parallel Hybrid Algorithm (PDPH), is proposed to improve the indexing performance for multi-core platform. Compared to existing sequential indexing algorithms, Pipeline and data parallelism are introduced by the PDPH to improve the algorithm flexibility and scale the performance with more cores. Evaluations showed this algorithm can improve index construction speed for multi-core platform.