National Repository of Grey Literature 2 records found  Search took 0.01 seconds. 
Comparison of Distributed File System Features
Humpa, Vítězslav ; Kašpárek, Tomáš (referee) ; Szőke, Igor (advisor)
The goal of this thesis is to compare parallel ditributed file systems. Thesis aims at GlusterFS, PVFS 2 and Gfarm FS. There is a description of each filesystem's theory, including general installation and configuration information. The text also informs about experimental work done with each system and evaluates test results which are then used to determine ways of filesystems' possible usage at our faculty. All tested filesystems have shown to be stable enough for use in one of the applications on FIT. As it's described in this thesis, GlusterFS would excel at applications that require simultaneous access to many large files (i.e. recorded lectures download system), all filesystems would give roughly the same performance when accessing many small files (i.e. webserver) and PVFS2 and Gfarm FS would be better suited for work with small number of large files (i.e. some distributed calculations).
Comparison of Distributed File System Features
Humpa, Vítězslav ; Kašpárek, Tomáš (referee) ; Szőke, Igor (advisor)
The goal of this thesis is to compare parallel ditributed file systems. Thesis aims at GlusterFS, PVFS 2 and Gfarm FS. There is a description of each filesystem's theory, including general installation and configuration information. The text also informs about experimental work done with each system and evaluates test results which are then used to determine ways of filesystems' possible usage at our faculty. All tested filesystems have shown to be stable enough for use in one of the applications on FIT. As it's described in this thesis, GlusterFS would excel at applications that require simultaneous access to many large files (i.e. recorded lectures download system), all filesystems would give roughly the same performance when accessing many small files (i.e. webserver) and PVFS2 and Gfarm FS would be better suited for work with small number of large files (i.e. some distributed calculations).

Interested in being notified about new results for this query?
Subscribe to the RSS feed.