Apriori-Based Prefetching Files for Caching

Authors

  • Prajwal Said UG – Information Technology Engineering, Bharati Vidyapeeth’s College of Engineering for Women, Dhankawadi, Pune, Maharashtra, India Author
  • Ketaki Naik UG – Information Technology Engineering, Bharati Vidyapeeth’s College of Engineering for Women, Dhankawadi, Pune, Maharashtra, India Author
  • Nupur Agrawal UG – Information Technology Engineering, Bharati Vidyapeeth’s College of Engineering for Women, Dhankawadi, Pune, Maharashtra, India Author
  • Srushti Bhoite UG – Information Technology Engineering, Bharati Vidyapeeth’s College of Engineering for Women, Dhankawadi, Pune, Maharashtra, India Author
  • Sayali Shelar Associate Professor – Information Technology Engineering, Bharati Vidyapeeth’s College of Engineering for Women, Dhankawadi, Pune, Maharashtra, India Author

DOI:

https://doi.org/10.47392/IRJAEM.2024.0339

Keywords:

Apriori, Caching, File Access Patterns, File System, Least Recently Used (LRU)

Abstract

The project proposes an innovative solution aimed at optimizing file system performance through predictive caching techniques integrated with a Graphical User Interface (GUI). The GUI facilitates user interaction by offering functionalities such as browsing files and displaying performance metrics via graphical representations of bandwidth and Input/Output Operations Per Second (IOPS). The functionality revolves around dynamically determining file placement on Solid State Drives (SSDs) or Hard Disk Drives (HDDs). The system employs predictive caching to identify frequently accessed files, ensuring faster retrieval by storing them on SSDs. Conversely, less frequently accessed files are allocated to HDDs. The project utilizes the Flexible I/O (fio) tool to measure the performance of files accessed on both SSDs and HDDs. Furthermore, to obtain insights and relationships between different files within the dataset, the project incorporates the Apriori algorithm. By analyzing structured relationships, the algorithm provides valuable intelligence for optimizing file placement decisions, enhancing overall system efficiency and adjust caching strategies to adapt to changing access patterns. By dynamically adapting file placement strategies based on access patterns and leveraging advanced algorithms for intelligent decision-making, the system endeavors to enhance user experience and system efficiency in managing file operations.

Downloads

Download data is not yet available.

Downloads

Published

2024-07-26