Azure App Service Plan中SocketOutboundAll指标聚合平均值计算咨询

Great question—let’s break down why you’re seeing such a stark difference between the Count and Average aggregations for the SocketOutboundAll metric:

1. What does SocketOutboundAll actually track? #

First, let’s clarify the metric’s core purpose: SocketOutboundAll captures the number of active outbound sockets at a specific moment in time for your App Service Plan. Each data point in this metric is a snapshot of that socket count during a regular sampling interval.

2. Why does Count aggregation show a stable value of ~1? #

The Count aggregation here doesn’t count the number of sockets—it counts how many metric data points are collected within your selected time window. If your chosen time granularity (e.g., 1 minute) matches the metric’s default sampling frequency, each window will only contain 1 snapshot data point. That’s why you’re seeing a consistent value of 1.

3. How is the Average aggregation calculated? #

The Average aggregation operates on an entirely different set of data:

• It takes the numeric value of every SocketOutboundAll data point in your selected time window (each value represents the active outbound socket count at that snapshot).
• It computes the arithmetic mean of those values.

So when you see an average between 40-50, that means during each aggregation window, the sampled active outbound socket counts averaged out to that range. This has no direct connection to the Count of data points—they’re measuring two completely separate things.

Quick Comparison #

内容的提问来源于stack exchange，提问作者Fraser