| Scyld ClusterWare HPC: Reference Guide | ||
|---|---|---|
| << Previous | Next >> | |
The node to query
A pointer to a statfs structure that will be filled upon successful completion. See the man page for statfs(2) for a description of the fields.
beostat_get_statfs_p will get the filesystem statistics for the root filesystem on a given node.
Warning: Since Linux aggressively caches the hard disk into memory it will often appear to always be about 90% used. Some have suggested that the values of buffers and cached added together should be subtracted from the reported memory usage. However, these values may not be mutually exclusive.
statfs_p_t statfs_p;
beostat_get_statfs_p (3, &statfs_p);
printf ("The node 3 has %s bytes free\n", statfs_p.mem.free); |
This function relies on the Beostat subsystem, which consists of the proc filesystem on the remote node, the sendstats daemon on the remote node, the recvstats daemon on the master node, and two shared memory files in the directory /var/shm. If any part of the system breaks down, this function could fail.
| << Previous | Home | Next >> |
| beostat_get_stat_cpu | Up | beostat_get_time |