Abstract—Distributed systems are computing systems where
a number of components cooperate by communicating over a
network. Few systems are still working in a stand-alone system
where the user interface, persistent data and applications
resided in one computer. However, the majority of nowadays
systems are designed to work in distributed systems.
Provide mechanisms for the storage and the manipulation of
large amount of data is one of the largest technological
challenges in software systems research today. Social media and
web services produce an impressive amount of data daily. In this
context, Hadoop Distributed file system (HDFS) is an open
source software framework for distributed storage and
processing of very large data (Big Data).
In this paper we introduce new functionalities for Hadoop
Distributed File System using probabilistic distributed
algorithms, our proposition is working in both homogenous and
heterogeneous HDFS nodes which can reduce the
communication cost.
Index Terms—Distributed systems, big data, Hadoop
distributed file system, probabilistic distributed algorithms.
The authors are with FS– Abdelmalek Essaâdi University, M’Hannech II
93030 Tetuan, Morocco (e-mail: ismailhind@gmail.com).
[PDF]
Cite: Ismail Hind and Ali Dahmani, "Toward a New HDFS Functionalities Using a Probabilistic Distributed Algorithm," International Journal of Future Computer and Communication vol. 4, no. 3, pp. 203-206, 2015.
