2013 | OriginalPaper | Chapter
Modeling and Simulation of Hadoop Distributed File System in a Cluster of Workstations
Authors : Longendri Aguilera-Mendoza, Monica T. Llorente-Quesada
Published in: Model and Data Engineering
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Considering the increased hard disk capacity on desktop PCs, we examine, by the modeling and simulation technique, the feasibility of exploiting the idle computational storage in a large Cluster of Workstations (COW). The model built is architecturally based on the Hadoop Distributed File System (HDFS) and was implemented in the CPN Tools using the Coloured Petri Nets combined with the CPN ML programming language. To characterize the workstations’ availability in the model, a statistical study was realized by collecting data from computer laboratories in our academic institution over a period of 40 days. From the simulation results, we propose a small modification in the source code of HDFS and a specific number of replicas in order to achieve a reliable service for writing and reading files despite the random failures due to the turning on and off of the computers in a COW with hundreds of machines.