NobleBlocks

Research indexJournal article

Hot Data Identification for Dynamic Workload using Parallel setup

Singh, Nerurkar, Pandat and 1 other

2022 IEEE Region 10 Symposium (TENSYMP)1 Jul 2022

NobleIDark:/48914/w1/20260515/99F19B95DOI10.1109/tensymp54529.2022.9864336

Paywalled

The full text is at the publisher

You can read the abstract here, or open it at the publisher.

Read at the publisher

Abstract

The use of Semantic Web applications is increasing day by day and so is the volume of Resource Description Framework (RDF) data. To manage the huge volume of RDF data, Hot Data identification for Dynamic Workload using parallel setup, HD2W is proposed. It exploits the skewness of the query workload. Data can be identified based on the frequency and recentness of the usage. It consists of two phases: Phase 1 generates the log and shuffles the data using randomized ranks and Phase 2 implements improved caching algorithm. The technique has been demonstrated using weather observation data collected by sensor networks. It uses a physical cluster for parallel implementation. The algorithm execution time increases linearly with workload. The experiments report 43% Query Execution Time (QET) gain over the original configuration. QET accelerates further for repeated occurrences. This work will help in developing interactive applications for the weather observation domain.

Discussion

Citations

Topics

More

Suggest a correction or report