What is a Memory Graph?
A Memory Graph is a visual representation of memory over time. FusionReactor periodically checks to see what the current state of the memory is and it will store these values in it's history. The type of graph you choose to view then effects how much of this memory you will see.
How do I view a Memory Graph?
From the FusionReactor Administrator, click "Current Memory Graph" or "Complete Memory Graph" from the table of contents (Memory section) and you will see the memory graph page. It will look something like this: (Although it is unlikely that the actual graph will look like this one.)
From this graph you can see the about of unallocated memory, the amount of allocated memory and the amount of memory actually being used. The first Current Memory Graph shows the most recent memory activity, whereas the Complete Memory Graph shows the memory activity over the whole of the stored history.
How do I configure the Memory Graph?
From the FusionReactor Administrator, click "memory Settings" from the table of contents (Memory section) and you will see the Memory Settings screen. It will look something like this:
From here you can change the way FusionReactor stores memory values:
Initial Delay: This is the number of milliseconds which FusionReactor will wait after a restart before collecting memory values. This wait is added so that you don't interfere with processes which must initialize after startup. It also avoids an unexpected spike at the start of the memory graph.
Sample Interval: This is the number of milliseconds which FusionReactor will wait between taking memory samples.
History Size: This is the maximum number of memory values which FusionReactor will keep in memory. Once the memory is full, the oldest value will be throw away every time a new one is taken.
The Current Memory Graph shows the last 300 memory samples. With a Sample Interval of 5000 milliseconds, the graph would show a time period of 25 minutes.
The Complete Memory Graph shows memory settings right through the history. With a History Size of 8640 and a Sample Interval of 5000 milliseconds, the graph would show a time period of 12 hours.
Why would I want to see a memory graph?
Because of its visual nature, the memory graph can very quickly give you a good idea how the system is performing. Not only that but, because it can cover larger periods of time, it can warn you of trends which could become a problem in the future. This kind of trend is much more difficult to spot if you are only checking the system periodically and are only looking at the health of the system right at this point in time.
