What are gauges?

Gauges capture a series of measurements where each one represents the value under observation at one point in time. The value of a gauge typically varies between some known minimum and maximum.

For example, concurrent TCP sessions on a given device is considered a gauge. There is no math done on a gauge object. It just takes the number polled and graphs that number.

Examples of gauge measurements include the requests/second serviced by an application, the amount of available disk space, the current value of $AAPL, etc.

With gauges, you have the option of submitting multiple summary statistics for a given metric as a single measurement using the form <min, max, sum, count>. This is useful for metrics that are sampled at a higher frequency than reported to Librato e.g. timing the latency of each request made to a web service. We also use this format when generating our historical aggregations at coarser resolutions (i.e. 1m, 15m, 60m).