Week 8: Cluster-based File Systems
- Lecture notes:
-
Cluster File Systems - Lecture slides
(6 per page)
- Supplemental notes:
-
- Google FS (GFS)
-
Sanjay Ghemawat, Howard Gobioff, and Shun-Tak Leung,
The Google File System ,
19th ACM Symposium on Operating Systems Principles, October, 2003.
- Hadoop Distributed File System (HDFS)
-
HDFS main page, hadoop.apache.org
HDFS Architecture Guide,
hadoop.apache.org
- Other stuff
-
Frank Schmuck and Roger Haskin,
GPFS: A Shared-Disk File System for Large Computing Clusters
.
Proceedings of the FAST 2002 Conference on File and Storage Technologies,
IBM Almaden Research Center San Jose, CA
-
xFS: Serverless Network File System,
University of California, Berkeley
- Buzzwords:
-
GFS
chunkserver, master, chunk handle, opeartion log, chunk lease, data flow, control flow
HDFS
NameNode, DataNode, blocks, FsImage
