Balance Page
A balanced cluster is one where resource usage is shared across all instances within the cluster, with no one instance overloaded.
Balance is calculated as a percentage from 0% (unbalanced) to 100% (fully balanced) and is based on a statistical comparison of 30 samples from a chosen metric. You can choose the metric used – logical reads, physical writes or CPU usage. Cluster Balance Metric
The charts in the Balance page display data for all three metrics, plus details of physical reads for the cluster.
To open the Balance page
- Select the Spotlight on Oracle RAC connection in the Spotlight Browser.
-
Click Cluster | Balance.
Charts on the Balance page
Notes:
- Every chart has a legend (list of symbols) to its right that describes the various series (line graphs) on the chart.
- Click an item in the legend to highlight its series (line) in the chart. Click a second time to return the series to its normal appearance.
- Move the mouse pointer over an item in the legend to view the current value for that series within the chart.
| Chart | Description |
|---|---|
| Logical Reads |
Logical Reads per Second is one of the ways in which Spotlight can measure balance between the instances in an Oracle RAC cluster. Each data series in the chart represents the rate of logical reads for a separate instance. A logical read occurs when Oracle finds data in the local buffer cache of an instance without needing to read the data from physical disk. The cluster is in balance when the rate of logical reads for all instances is approximately the same. |
| CPU Usage |
CPU Usage is one of the ways in which Spotlight can measure balance between the instances in a cluster. Each data series in the chart represents the CPU usage for a separate instance. The cluster is in balance when the CPU usage across all instances is approximately the same. |
| Physical Reads |
Physical Reads per Second is provided for comparison with the Logical reads chart above. Each data series in this chart represents the rate of physical reads for a separate instance. A physical read occurs when a database instance retrieves data from datafiles on the shared disk subsystem, and occurs every time a logical read fails to find the requested page in local cache or global cache. |
| Physical Writes |
Physical Writes per Second is one of the ways in which Spotlight can measure balance between the instances in a cluster. Each data series in the chart represents the rate of physical writes for a separate instance. A physical write occurs every time a database instance writes data to a file on the cluster's shared disk sub-system. The cluster is in balance when the rate of physical writes from all instances is approximately the same. |