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

# Replication delay metric in ELT & CDC

> How the Replication Delay metric measures the time since data was last copied from S3 to the destination in Integrate.io ELT & CDC.

## Overview

The **Replication Delay** metric measures the time, in minutes, since the pipeline last copied data from the intermediary storage (S3) to the destination. It provides a clear indicator of how timely data replication is occurring.

This metric is available for pipelines with **Redshift** or **Snowflake** as the destination, regardless of the source type.

## Key Features

* **Replication Delay Monitoring**:
  Measure the time elapsed since the last replication from the pipeline's intermediary storage to the destination. This helps ensure data replication occurs as expected.
* **Alert Setup**:
  Set up alerts to notify you if the replication delay exceeds the expected sync frequency time. For example, if a pipeline’s sync frequency is 10 minutes, an alert threshold of 11 minutes can indicate potential pipeline issues.

## Considerations

* **Pipeline State**:
  Alerts are triggered only for pipelines in the **Running** state. Metrics will not be available for pipelines in other states.
* **Initial Sync Behavior**:
  If the sync has just started and the alert runs before the first replication to the destination, the metric will return a value of **0**.
* **Metric Scope**:
  The Replication Delay metric measures the delay between the **pipeline (intermediary storage)** and the **destination**. It does **not** measure the delay between the source and the intermediary storage. For monitoring the delay between the source and the pipeline, use the [**Sync Lag Metric**](/cdc/available-metrics-types-sync-lag-metric).