[Yandex Cloud documentation](../../index.md) > [Yandex Cloud Detection and Response](../index.md) > Concepts > About Yandex Cloud Detection and Response

# Yandex Cloud Detection and Response overview

{% note info %}

This feature is in the [Preview](../../overview/concepts/launch-stages.md) stage. To get access, contact [tech support](https://center.yandex.cloud/support) or your account manager.

{% endnote %}

Yandex Cloud Detection and Response provides incident monitoring and response in Yandex Cloud. YCDR is built around Yandex Cloud's in-house Security Operations Center (SOC). This service collects data from the cloud infrastructure to detect anomalies. When YCDR detects an anomaly, it creates an [alert](../../security-deck/concepts/alerts.md) indicating a potential incident.

## Access to the service {#access}

{% note info %}

You get access to YCDR in the Security Deck interface in [Cloud Center](https://center.yandex.cloud) after your access request gets approved.

{% endnote %}

The Yandex Cloud SIEM system analyzes the collected data. Events are sent to the SIEM system via a _collector_. The collector is installed in a Managed Service for Kubernetes cluster, which ensures its scalability and fault-tolerance.

The collector must have access to the external network to send events to the Yandex Cloud SIEM. Yet, since events are sent over the TLS protocol and SIEM is physically located in the Yandex Cloud infrastructure, the data remains inside the data center.

The collector works at the [cloud](../../resource-manager/concepts/resources-hierarchy.md) level. Each cloud must have a dedicated collector for sending events.

The collector architecture includes the following modules:

1. `Vector`-based component for collecting and sending events. It enables receiving events from `osquery` agents and random events over HTTP.
1. `syslog` event collection component which collects events and sends them to the `Vector`-based component for further processing.

In Yandex Cloud Detection and Response, you can access a list of detected incidents and select one to get troubleshooting recommendations with additional context and view the incident details and category. To see the statistics on detected incidents, refer to the dashboard on the service's home page.