> For the complete documentation index, see [llms.txt](https://docs.seldon.ai/seldon-core-2/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.seldon.ai/seldon-core-2/user-guide/scaling/hpa-overview.md).

# Using HPA for Autoscaling

Given Seldon Core 2 is predominantly for serving ML in Kubernetes, it is possible to leverage `HorizontalPodAutoscaler` or [HPA](https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/) to define scaling logic that automatically scales up and down Kubernetes resources. HPA targets Kubernetes or custom metrics to trigger scale-up or scale-down events for specified resources. Using HPA is recommnended if **custom Scaling Metrics** are required. These would be exposed using Prometheus, and [Prometheus Adapter](https://github.com/kubernetes-sigs/prometheus-adapter) or similar tools for explosing metrics to HPA. If these tools cause conflicts, [autoscaling functionality native to Core 2](/seldon-core-2/user-guide/scaling/core-autoscaling.md) that does not require exposing custom metrics is recommended.

Seldon Core 2 provides two main approaches to leveraging Kubernetes Horizontal Pod Autoscaler (HPA) for autoscaling. It is important to remember that since in Core 2 Models and Servers are separate, autoscaling of *both* Models and Servers, in a coordinated way, needs to be accounted for when implementing autoscaling. In order to implement either approach, metrics first need to be exposed - this is explained in the [HPA Setup](/seldon-core-2/user-guide/scaling/hpa-overview/hpa-setup.md) guide which explains the fundamental requirements and configuration needed to enable HPA-based scaling in Seldon Core 2.

## 1. Model Autoscaling with HPA

The [Model Autoscaling with HPA](/seldon-core-2/user-guide/scaling/hpa-overview/model-hpa-autoscaling.md) approach enables users to scale Models based on custom metrics. This approach, along with [Server Autoscaling](/seldon-core-2/user-guide/scaling/core-autoscaling/core-autoscaling-servers.md), enables users to customize the scaling logic for models, and automate the scaling of Servers based on the needs of the Models hosted on them.

![Model Autoscaling with HPA, Servers autoscaled by Core 2](/files/jQU3LJeKrXbTAQgmIZqM)

## 2. Model and Server Autoscaling with HPA

The [Model and Server Autoscaling with HPA](/seldon-core-2/user-guide/scaling/hpa-overview/single-model-serving-hpa.md) approach leverages HPA to autoscale for Models and Servers in a coordinated way. This requires a **1-1 Mapping of Models and Servers** (no Multi-Model Serving). In this case, setting up HPA can be set up for a Model and its associated Server, targetting the same *custom* metric (this is possible for Kubernetes-native metrics).

![Model and Server autoscaling with HPA, for single-model serving](/files/4kHHkcknCB6jStq3sIuK)

s


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.seldon.ai/seldon-core-2/user-guide/scaling/hpa-overview.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.