Multithreaded Variable Chunking in Source Based Deduplication in Cloud Backup Services using Support Vector Machine

Abstract

Cloud computing is typically defined as a type of computing that relies on sharing computing resources rather than having local server or personal devices to handle applications. we developed a novel multithreaded variable size chunking method,which exploits the multicore architecture of the modern microprocessors. The legacy single threaded variable size chunking method leaves much to be desired in terms of effectively exploiting the bandwidth of the state of the art storage devices. Multithreaded variable size chunking guarantees chunking invariability: The result of chunking does not change regardless of the degree of multithreading or the segment size. This is achieved by inter and intra-segment coalescing at the master thread and Dual Mode Chunking at the client thread. Then developed an elaborate performance model to determine the optimal multithreading degree and the segment size. MUCH is implemented in the prototype deduplication system. By fully exploiting the available CPU cores (quad-core), This achieved up to 4 increase in the chunking performance (MByte/sec).Multithreaded variable size chunking successfully addresses the performance issues of file chunking which is one of the performance bottlenecks in modern deduplication systems by parallelizing the file chunking operation while guaranteeing Chunking Invariability by using Support Vector Machine.

Authors and Affiliations

K. S. Demel Abisha, S. Revathy Rajesh

Keywords

Related Articles

A Proposed approach of VANET using LI Technique for safe city and vehicle

VANET is an important technique that provides communication between vehicles over the wireless link. There are many protocol proposed for VANET that provide many type of the services to the network but they cannot provi...

Enrichment Analysis for Gene Dataset

Enrichment Analysis is carried out for gene expression analysis. Many genes do not obey the central dogma, there are variations in level of expression at each step of the central dogma. Upregulated and downregulated gen...

Onset Instability of Thermosolutal Convective Flow of Viscoelastic Maxwell Fluid Thorough Porous Medium with Linear Heat Source Effect

The Instability mechanism of Thermosolutal convection in a horizontal layer of viscoelastic Maxwell fluid through porous medium with internal linear heating is presented in this manuscript. The flow is also affected wit...

Granular Computing for Data Mining

Granular computing is a rising computing worldview of data handling. It concerns the handling of complex data substances called data granules, which emerge during the time spent information reflection and induction of l...

Study of Particle Swarm Optimization Based Interconnected Automatic Generation Control System

This paper with help of controllers considering equal area to minimize errors in power system. The areas are equal in nature. The sources used here depicts with the automatic generation control of multisource interconne...

Download PDF file
  • EP ID EP21991
  • DOI -
  • Views 220
  • Downloads 5

How To Cite

K. S. Demel Abisha, S. Revathy Rajesh (2016). Multithreaded Variable Chunking in Source Based Deduplication in Cloud Backup Services using Support Vector Machine. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 4(4), -. https://europub.co.uk/articles/-A-21991