Back to Articles

A request to graph KPIs for a large number of modems over a long period of time, such as the 14-day maximum, can result in a dataset that is too large to graph. ServAssure NXT solves this problem by running a pruning algorithm on the data returned before passing it to the ServAssure NXT User Interface for display in a graph. This algorithm examines multiple data points for differences. If there are no differences between the data points, or the differences are small, ServAssure NXT eliminates a data point and simplifies the series line. A data point is only eliminated if the difference between it and the previous data point is very small. This data pruning permits ServAssure NXT to graph and display data collected for a large number of cable modems, at short intervals, over long periods of time.

Data pruning occurs for all cable modem KPI graph displays, but does not occur for Cable Modem Status or Utilization KPIs. Cable Modem Status graph lines are enumeration states that indicate overall connectivity to the network, not percentages or measurements. Utilization graph lines show all data points in the series, to permit detailed evaluation of critical channel performance indicators. Evidence of data pruning is most noticeable when viewing a graph of Utilization values next to a graph of non-Utilization KPI values. The Utilization graph will show more data points per line when compared to the graph of non-Utilization KPIs, as illustrated in the image below.