About this document

Big Data: Google File System Overview by Abhay Dabhade is a document available to read on EtoBox.

The document discusses the Google File System (GFS) and Hadoop Distributed File System (HDFS). GFS uses a master/slave architecture with one master node and multiple chunkservers. Files are divided into 64MB chunks which are replicated across servers for fault tolerance. The master manages metadata like file mappings and chunk locations. HDFS also uses this master/slave model with a Namenode as master and Datanodes as slaves, and also replicates blocks of data for reliability.

Author

Abhay Dabhade

Language

EN