Performance issue when clustered nodes access same file via NFS
Issue
-
6 nodes of 3rd party cluster using custom application. Each instance reads through a file on the NFS server and looks for certain parts of the file. These parts are then written to memory and in the end they are written to a new file. Size of the input file is about 20-25GB and each output file is about 300MB.
-
The process is slow only when it runs across many machines. If running on a single machine it's not slow.
Environment
- Red Hat Enterprise Linux 6
- NFSv4
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.