Article: Workload aware incremental repartitioning of NoSQL for OLTP applications Journal: International Journal of Internet of Things and Cyber-Assurance (IJITCA) 2020 Vol.1 No.3/4 pp.214 - 231 Abstract: Numerous applications are deployed on the web with the increasing popularity of internet. These include gaming and e-commerce web applications. These applications generate huge amount of the data. One particular machine cannot handle such data. It is difficult to scale out by using relational databases. OLTP system needs to be scalable and require fast response. Therefore the scalability becomes the challenge for the e-commerce applications. Data partitioning technique is used to improve the scalability of the system. Existing partitioning techniques does not consider the relation among tuples. These techniques do not handle incremental data and are suitable for those applications that required only sequential access to the data. It results in increasing the number of the distributed transactions. The work-load aware incremental repartitioning approach is used to balance the load among the partitions. Hypergraph representation technique represents transactional workload in graph form. In this technique, frequently used items are grouped together by using fuzzy C-means clustering algorithm. Tuple classification and migration algorithm is used for mapping clusters to partitions and after that tuples are migrated efficiently. Inderscience Publishers - linking academia, business and industry through research

Title: Workload aware incremental repartitioning of NoSQL for OLTP applications

Authors: Anagha Bhunje; Swati Ahirrao

Addresses: Symbiosis Institute of Technology, Symbiosis International University, Pune, India ' Symbiosis Institute of Technology, Symbiosis International University, Pune, India

Abstract: Numerous applications are deployed on the web with the increasing popularity of internet. These include gaming and e-commerce web applications. These applications generate huge amount of the data. One particular machine cannot handle such data. It is difficult to scale out by using relational databases. OLTP system needs to be scalable and require fast response. Therefore the scalability becomes the challenge for the e-commerce applications. Data partitioning technique is used to improve the scalability of the system. Existing partitioning techniques does not consider the relation among tuples. These techniques do not handle incremental data and are suitable for those applications that required only sequential access to the data. It results in increasing the number of the distributed transactions. The work-load aware incremental repartitioning approach is used to balance the load among the partitions. Hypergraph representation technique represents transactional workload in graph form. In this technique, frequently used items are grouped together by using fuzzy C-means clustering algorithm. Tuple classification and migration algorithm is used for mapping clusters to partitions and after that tuples are migrated efficiently.

Keywords: online transaction processing; OLTP; distributed transactions; not only SQL; NoSQL; incremental repartitioning technique; hyper graph; scalability; fuzzy C-means clustering algorithm.

DOI: 10.1504/IJITCA.2020.112511

International Journal of Internet of Things and Cyber-Assurance, 2020 Vol.1 No.3/4, pp.214 - 231

Received: 22 May 2017
Accepted: 19 Nov 2017
Published online: 20 Jan 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Workload aware incremental repartitioning of NoSQL for OLTP applications

Keep up-to-date