To smooth a very spiky cpu.utilization metric, what is the correct analytic function to better see if the cpu. utilization for servers is trending up over time?

Question

Options

• ARate/Sec
• BMedian
• CMean (by host)
• DMean (Transformation)

• A
  6% (2)
• B
  8% (3)
• C
  14% (5)
• D
  72% (26)

Explanation

According to the web search results, a mean transformation is an analytic function that returns the average value of a metric or a dimension over a specified time interval. A mean transformation can be used to smooth a very spiky metric, such as cpu.utilization, by reducing the impact of outliers and noise. A mean transformation can also help to see if the metric is trending up or down over time, by showing the general direction of the average value. For example, to smooth the cpu.utilization metric and see if it is trending up over time, you can use the following SignalFlow code: mean(1h, counters("cpu.utilization")) This will return the average value of the cpu.utilization counter metric for each metric time series (MTS) over the last hour. You can then use a chart to visualize the results and compare the mean values across different MTS.

No community discussion yet for this question.