Optimizing OLAP heterogeneous computing based on Rabin-Karp algorithm
The enormous amount of data has been boundlessly growing over the last few decades and expected to exponentially do so in the future. However, a substantial size of this accumulated amount is discarded anyhow. The processing capabilities have been considered as one of the major barriers in the...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/33185/ http://irep.iium.edu.my/33185/ http://irep.iium.edu.my/33185/1/Optimizing_OLAP_Heterogeneous_Computing_Based.pdf |
| Summary: | The enormous amount of data has been boundlessly
growing over the last few decades and expected to exponentially
do so in the future. However, a substantial size of this
accumulated amount is discarded anyhow. The processing
capabilities have been considered as one of the major barriers in
the way of exploiting this priceless mine. Therefore, the issue has
absorbed considerable part of researchers’ concentration. OLAP
has been considered as a powerful method for analysing
excessive size of data that works closely to intelligent business,
medical fields. Yet, such a method will always need increasing
processing resources. Numerous enhancements have been
suggested in order to improve OLAP performance, part of them
has gone to the processing capabilities whereby parallel
processing has occupied a sizeable space. A heterogeneous
computing is a relatively recent approach that is being under
examination. In this paper, through experimental results and
based on Rabin-Karp Algorithm; we propose an optimized
heterogeneous solution that takes into account the benefits and
the boundaries in order to achieve a better OLAP performance
in terms of response time with three times gain. In the light of
our results; we present the achieved gain and possible future
trends.
Keywords— OLAP, heterogeneous computing, Rabin-Karp,
data mining, pattern recognition. |
|---|