eG Monitoring
 

Measures reported by NFSThreadUtilTest

When the Atlantis ILIO receives an I/O request, an NFSD (Network File System Daemon) thread is generated to handle the client request for the file system operation. Each NFSD thread can process one request at a time. A large pool of threads can allow the server to handle more NFS requests in parallel. Also, the daemons are self-tuning i.e., they can create or delete the threads as needed, based on the amount of NFS activity. The receipt of any NFS protocol request from the client requires the dedicated attention of the nfs daemon thread until that request is satisfied and the results of the request processing are sent back to the client. Often, adequate threads may not be available to process the requests thus causing a processing bottleneck which when left unattended will have a considerable impact on the performance of the Atlantis ILIO. Therefore, it is imperative to monitor the NFS threads so that administrators can proactively be alerted to performance bottlenecks. The NFS Thread utilization test exactly adminsitrators in this process!

This test reports the number of threads that are currently executing on the Atlantis ILIO and the number of times all the threads have been utilized by the Atlantis ILIO.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Threads Indicates the total number of NFS threads that are currently executing on this server. Number This measure is a good indicator of the level of NFS activity on the server.
Thread_in_use Indicates the number of times all the NFS threads were in use. Number A low value is desired for this measure. A high value indicates that the I/O requests sent to the NFS datastore from the server are blocked and are waiting for reply.