L.9 Processing time required for configuring and clearing monitoring settings
This subsection describes how to estimate the amount of time required for configuring monitoring settings after a scale out is detected by the manager host, and the amount of time required for clearing monitoring settings after a scale in is detected.
- Organization of this subsection
(1) Time required for configuring monitoring settings after a scale out
Use the formulas in the table below to estimate the amount of time required for configuring the monitoring settings of the manager host for agent hosts that are added during a scale out.
OS |
Processing time for configuring monitoring settings |
---|---|
Windows |
((Number of hosts managed in the configuration before scale out + number of hosts managed in the configuration after scale out)/2) x (15 seconds/100 hosts) x number of hosts added during scale out |
Linux |
((Number of hosts managed in the configuration before scale out + number of hosts managed in the configuration after scale out)/2) x (5 seconds/100 hosts) x number of hosts added during scale out |
- Note
-
In a large-scale environment where the configuration of many hosts (600 hosts or more) is managed, we recommend that you limit the number of hosts that can be scaled in or out to 20 hosts per hour.
(2) Time required for clearing monitoring settings after a scale in
Use the formulas in the table below to estimate the amount of time required for clearing the monitoring settings of the manager host for agent hosts that are deleted during scale in. Note that this processing time assumes that the default value NO is set for the setting parameter JbsrtDistribByScaleIn.
OS |
Processing time for clearing monitoring settings |
|
---|---|---|
When JP1/PFM is used |
When JP1/PFM is not used |
|
Windows |
(About 35 seconds per host) x number of hosts deleted during scale in |
(About 6 seconds per host) x number of hosts deleted during scale in |
Linux |
(About 30 seconds per host) x number of hosts deleted during scale in |
(About 5 seconds per host) x number of hosts deleted during scale in |