AnalyzingGFS_HDFS
AnalyzingGFS_HDFS
net/publication/308281665
CITATIONS READS
7 2,306
1 author:
Nader Gemayel
Notre Dame University
2 PUBLICATIONS 33 CITATIONS
SEE PROFILE
All content following this page was uploaded by Nader Gemayel on 30 July 2020.
Research Article
Analyzing Google File System and Hadoop Distributed File System
Nader Gemayel
Department of Compuer Science, Notre Dame University-Louaize, P.O. Box 72, Zouk Mosbeh, Zouk Mikayel, Lebanon
Control message File name, chunk index GFS master server
Application
Data message
GFS client File name space
Chunk location
Root /root/dir 1
Chunk 5f 3s
Dir 1
Dir 2 Chunk 7f 2a
Chunk data Chunk location byte range
Chunk 8d3a
Data connection
Management
Zookeeper Chukwa
Data access
Pig Hive Avro
Data processing
MapReduce framework
Data storage
Hadoop Distributed File
System (HDFS) HBase
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C C
C
C
C
Control message
Data message
Write data
Chunkserver X Chunk
Buffer Block A
Control message
Data message
1 Permission to write “Hello.txt”
HDFS client NameNode
Write block A to DataNode: 1, 7, 9 2
3 Rack awareness:
Ac
k:
Re
Rack 1
B
ad
lo
Rack 2
ce
iv
Data node 7
ed
Data node 9
7
Ye
s:
Re
Switch 1 Switch 2
ed
eiv
ad
File: Hello.txt
y?
rec
Block A Block B Block C Data node 1 Data node 7
A
ck
Yes, ready? Block A
5 Write block A Blo
Block A
k:
Ac
4 Yes, ready? 6
Data node 9
Block A
Rack 1 Rack 2
C C
C
C
View publication stats