Help

Table Of Contents
often typical of back-ups or batch processing jobs. These spikes are not necessarily a sign of a problem. However, monitor their
height and duration, and their impact on response time.
NOTE: The chart displays data only for the directors in the cluster to which you are currently connected. To simultaneously
view statistics for another cluster, open a second browser session and connect to the second cluster.
Guidelines
Keep in mind the following guidelines when using the CPU Utilization chart:
A completely idle system will be around 0-5% busy
Less than 50% CPU utilization is safe
CPU utilization of 50-80% warning
CPU utilization of 80-90% danger
CPU utilization of >90% critical
Corrective actions
Consider migrating virtual volumes to less busy directors (moving them to different storage views).
Consider adjusting the host failover driver settings to better load balance I/O across available directors.
Stagger batch application start times causing troublesome CPU spikes (such as modify your back-up jobs schedule to start
with some delay between jobs instead of starting all back-up jobs at once, if possible.)
Look for signs of perhaps a rogue director firmware process or thread hogging CPU time.
If normal baseline busy is constantly high (due to high I/O load), consider an engine expansion, or a VS1 to VS2 hardware
upgrade, if possible.
Changing the view
To view the CPU utilization of a single director in your metro node system, select the director name from the Director
drop-down.
Viewing the CPU Utilization chart
1. From the GUI main menu, click Performance.
2. In the Performance Dashboard, select the tab in which you want to display the CPU Utilization chart (or create a custom
tab).
3. Click +Add Content.
4. Click the CPU Utilization chart icon.
Heap Usage chart
The Heap Usage chart shows a percentage of the heap memory used by the firmware on a director. In GeoSynchrony 5.x, the
heap size is 6GB.
The greater the number of 4KB pages storing user data that the metro node director cache needs to keep track of, the greater
the heap usage. There are safeguards in place so that the amount of information recorded about pages known by the cache will
not exceed the heap size. If the amount of information recorded gets too large, the cache layer will remove the cached pages
and knowledge about them from it's cache.
During failure recovery, the amount of heap usage initially increases, and then decreases back to normal levels once the failure is
handled. For this reason, it's important that a director's heap not get too full (< 85-90%) to account for this.
Guidelines
Keep in mind the following guidelines when using the Heap Usage chart:
A value of around 40 to 50% is normal.
Monitoring the system
37