The “Trend View” feature that allows you to see a graphical representation of data points on a monitor over a period time.
By selecting feature you will see a visual representation that depicts the average (or general) tendency of a series of data points to move in a certain direction during a certain time, and is drawn on the monitor as either a straight line or a curve, depending which option you select.
This feature can be utilized on all uptime monitors. It is a “one time” feature, meaning that when you select it the trend line will appear (if there is not sufficient data then no Trend Line will appear) and will be automatically deselected after 5 minutes (as your monitors refresh) when data will be reloaded. If you wish to see it again you need to press the appropriate button under the monitor.
Please see the examples below.
The “Average data” on the monitor is the average of the accumulated data from the selected locations for a given period of time. It represents server performance (for example, average response time) based on each locations information.
In most cases data is too noisy; therefore we apply a smoothing/fitting technique so that the information is easier to understand and thus more useful.
Linear trends show either a decrease or increase of average values over time. It is a very simple line fitting, i.e. a linear two point graph. The first point is the average best fit value at start time of given interval and the second point is the average best fit value at end time of given interval.
So these two values just show a straight line graph from start to finish and ignore fluctuations that happen between the start point and the end point.
The “smoothing” option provides a more detailed analysis of data behavior. Data is divided into defined subsets and a linear trend is calculated for each subset.
The “spline” gives you a depiction of overall data behavior while still cleaning out random spikes and fluctuations.
Spline (another smoothing technique) is a polynomial image that is obtained from the subsets of values of the original data.