# Deka GPU: Autoscaling

- [Basic Autoscaling](/deka-gpu/deka-gpu-autoscaling/basic-autoscaling.md)
- [KEDA Autoscalling](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling.md)
- [Introduction](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/introduction.md)
- [Install KEDA](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/install-keda.md): To install KEDA, you need to create a namespace and a VPC first by running the following commands.
- [Add the Helm Repository](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/add-the-helm-repository.md)
- [Create values.yaml from Helm Repository](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/create-values.yaml-from-helm-repository.md)
- [Example: Autoscaling vLLM with KEDA based on GPU KV Cache Usage](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/example-autoscaling-vllm-with-keda-based-on-gpu-kv-cache-usage.md)
- [Introduction](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/example-autoscaling-vllm-with-keda-based-on-gpu-kv-cache-usage/introduction.md)
- [Metric: GPU KV Cache Usage](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/example-autoscaling-vllm-with-keda-based-on-gpu-kv-cache-usage/metric-gpu-kv-cache-usage.md)
- [KEDA Configuration Example](/deka-gpu/deka-gpu-autoscaling/keda-autoscalling/example-autoscaling-vllm-with-keda-based-on-gpu-kv-cache-usage/keda-configuration-example.md)