content/commands/latency-histogram.md
LATENCY HISTOGRAM returns a cumulative distribution of commands' latencies in histogram format.
By default, all available latency histograms are returned. You can filter the reply by providing specific command names.
Each histogram consists of the following fields:
This command requires the extended latency monitoring feature to be enabled, which is the default.
If you need to enable it, call CONFIG SET latency-tracking yes.
To delete the latency histograms' data use the [CONFIG RESETSTAT]({{< relref "/commands/config-resetstat" >}}) command.
127.0.0.1:6379> LATENCY HISTOGRAM set
1# "set" =>
1# "calls" => (integer) 100000
2# "histogram_usec" =>
1# (integer) 1 => (integer) 99583
2# (integer) 2 => (integer) 99852
3# (integer) 4 => (integer) 99914
4# (integer) 8 => (integer) 99940
5# (integer) 16 => (integer) 99968
6# (integer) 33 => (integer) 100000
| Redis Software | Redis Cloud | <span style="min-width: 9em; display: table-cell">Notes</span> | |:----------------------|:-----------------|:------| | <span title="Supported">✅ Standard</span> <span title="Supported"><nobr>✅ Active-Active</nobr></span> | <span title="Supported">✅ Standard</span> <span title="Supported"><nobr>✅ Active-Active</nobr></span> | |
{{< multitabs id="latency-histogram-return-info" tab1="RESP2" tab2="RESP3" >}}
Array reply: a map where each key is a command name, and each value is a map with the total calls, and an inner map of the histogram time buckets.
-tab-sep-
Map reply: a map where each key is a command name, and each value is a map with the total calls, and an inner map of the histogram time buckets.
{{< /multitabs >}}