Owners Manual

Best Practices: Performance and Monitors | Performance Monitoring
442 OMNM 6.5.2 User Guide
system would then only have an insert rate of 254 per second well below the hardware’s
limitation.
NOTE:
Traffic flow analysis can process and retain even larger amounts of information. Flows that correlate
50%, polled every minute for a day require roughly 109G of database, and require 4500 insertions per
second.
CAUTION:
These numbers and sample calculations represent best case scenarios. Any disk or disk array typically
serves other applications and processes besides monitors. Make sure to take account of that when
calculating how to accommodate your monitors. The system admin or system user should assist in
making that assessment.
Storage Requirements (Database Size) — How much disk space do you need, based on your
retention policy? See
Retention Policies on page 376 for more about configuring those.
OpenManage Network Manager stores performance data in three different forms Detail,
Hourly and Daily data. It collects Detail metrics directly from the device or calculates these
from the collected data with each poll. Hourly data summarizes the detail data collected
during the hour. Daily data summarizes the hourly data collected during that day. The
retention policy associated to the monitor describes how long OpenManage Network
Manager retains each of these data types within the system.
OpenManage Network Manager stores a performance metric as a single row within a database
table. Each row in that database consumes roughly 150 bytes of disk space. The sum of each
monitor’s disk space required determines the amount of disk space. For each monitor add the
disk space required for the Detail, Hourly and Daily data using the following formula:
<Detail disk space> + <Hourly disk space> + <Daily disk space> = Monitor disk space in
bytes
where...
<# of metrics retained per poll> = <# of monitor targets> x <# of retained attributes>.
<# of metrics retained per day> = <# of metrics retained per poll> x <# of polls per
day>.
<Detail disk space> = <# of metrics retained per day> x <# of days to retain Detail data>
x 150.
<Hourly disk space> = <# of metrics retained per poll> x <# of days to retain Hourly
data> x 24 x 150.
<Daily disk space> = <# of metrics retained per poll> x <# of days to retain Hourly data>
x 150.
If the system does not have sufficient disk space consider the following options:
• Add more hardware to increase the available disk space.
• Reduce the retention period of one or more monitors to lower the overall disk space
requirements. Of the three data forms Detail data will consume the most disk space per day
of retention.