Listing of Tables
attributes table (see Section 10.2.3) must be set to cpu. This is the default setting.
The interval at which the nodes are sampled for CPU statistics is controlled by the
attribute cpu-stats-poll-intervalin the attributes table; the default is to
sample every 2 minutes.
The node statistics (node_stats) table can grow rapidly,especially on a large machine.
Running the table administration program, rmstbladm, with the -c option removes old
entries. This should be done periodically using a cron script. See Page 5-44 for details.
The stats-lifetimeattribute in the attributes table (see Section 10.2.3)
determines how oldthe entries must be before they are removed.
Table10.21: Node Statistics Table
Name Type Description
name char(16) name of the node
ctime UTC time at which sample was collected
usercpu % user CPU time since last sample
syscpu % system CPU time since last sample
freemem int free memory in megabytes
ubc int size of the unified buffer cache in megabytes
wired int wired memory in megabytes
freeswap int free swap space in megabytes
pages int page fault rate
interrupts int interrupts rate (except clock)
syscalls int system call rate
users int number of users logged in
freetmp int free temporary file space in megabytes
The usercpu and syscpu statistics are percentages calculated over the periodsince the
last sample.
The interrupts,pages and syscalls statistics are rates averaged over the interval
since the last sample.
10.2.16 The Partitions Table
The nodes in a machine aregrouped into partitions according to their function. For
example,there maybe an administrative partition, a login partition and a parallel
programming partition. A set of partitions spanning the machine is called a
configuration. Different configurations may be appropriate to different times of the day
or week. For example, one for daytime running and another for nights and weekends.
Only one configuration, the active configuration, can be running at a time.
The RMS Database 10-17