> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vortexiq.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Command Latency p50 (us), Redis

> Command Latency p50 (us) for Redis instances. Tracked live in Vortex IQ Nerve Centre. How to read it, why it matters, and how to act on it.

**Card class:** [Non-Hero](/nerve-centre/overview#card-classes-explained)  •  **Category:** [Performance](/nerve-centre/connectors#connectors-by-type)

## At a glance

> **Command Latency p50 (us)** is the median time, in microseconds, that the Redis server spends executing a command, measured server-side rather than including network round-trip. Healthy Redis sits comfortably in the tens-to-hundreds of microseconds at the median, so this card is your baseline "is the engine itself fast?" reading. It is the calm-weather companion to the p95 and p99 latency cards: when p50 itself starts climbing, the slowdown is broad rather than tail-only.

|                    |                                                                                                                                                                  |
| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **What it tracks** | The 50th-percentile (median) server-side command execution latency in microseconds (us).                                                                         |
| **Data source**    | `LATENCY HISTORY` (Redis 7+ per-event latency tracking) where available, or client-side sampling of command round-trips when the latency monitor is not enabled. |
| **Time window**    | `RT/5m` (real-time, rolled over a trailing 5-minute window).                                                                                                     |
| **Alert trigger**  | None set. p50 is a baseline-watch metric, not a paging metric; the tail-latency cards carry the alert thresholds.                                                |
| **Roles**          | engineering, operations                                                                                                                                          |

## What it tracks

Redis is single-threaded for command execution, so the time the server spends inside a command is a direct, uncontended measure of how much work each command costs. The median (p50) tells you what a "typical" command looks like right now: a `GET` against a small string or a `HGET` against a modest hash should land in the low tens of microseconds. A rising p50 means the typical command is getting more expensive, which usually points at growing value sizes, an O(N) access pattern creeping into hot paths, or CPU contention on the host. Because p50 ignores the slow tail entirely, it is the cleanest signal that a regression is systemic rather than confined to a few pathological keys. Reported in microseconds because well-behaved Redis commands are sub-millisecond; surfacing this in us keeps the everyday range readable instead of rounding everything to "0ms".

## Reconciling against the source

Confirm the reading against Redis's own latency tooling: run `LATENCY HISTORY command` (Redis 7+) or `LATENCY LATEST` for the per-event monitor, and `redis-cli --latency` / `--latency-history` for an end-to-end sampled view. On ElastiCache or MemoryDB, cross-check the `SuccessfulReadRequestLatency` and `SuccessfulWriteRequestLatency` CloudWatch metrics; note those are microsecond engine-side timings too, so they should track this card closely once the period and node are matched.

***

### Tracked live in Vortex IQ Nerve Centre

*Command Latency p50 (us)* is one of hundreds of KPI pulses Vortex IQ tracks across Redis and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English.

[Start for free](https://app.vortexiq.ai/login) or [book a demo](https://www.vortexiq.ai/contact-us) to see this metric running on your own data.