[Yandex Cloud documentation](../../index.md) > [Yandex Managed Service for Kubernetes](../index.md) > [Concepts](index.md) > Recommended master configurations

# Recommended master configurations

You can select [master](index.md#master) configuration when [creating](../operations/kubernetes-cluster/kubernetes-cluster-create.md) or [updating](../operations/kubernetes-cluster/kubernetes-cluster-update.md#manage-resources) your cluster. For stable operation, make sure the master configuration matches the workload of your Kubernetes cluster.

You can use this table to find the minimum master configuration based on your specific cluster specifications. Configurations may change over time. You can view the list of available configurations [here](../operations/kubernetes-cluster/kubernetes-cluster-configuration-list.md).

The selected configuration allocates minimum resources to the master. If you enable [master node auto-scaling](autoscale.md#master-autoscaler) in your cluster, your selected configuration will serve as the lower scaling limit.

{% note warning %}

Starting June 18, 2026, master node auto-scaling will be enabled on all clusters in the `RAPID` [release channel](release-channels-and-updates.md) where the master is deployed in a [highly available configuration](index.md#master).

{% endnote %}

#|
|| **Cluster specifications** {align="center"} | > | > | **Estimated master node requirements** {align="center"} | > | **Recommended master configuration** {align="center"} | > | > ||
|| **CNI** | **Nodes** | **Maximum number of pods** | **vCPU** | **RAM** | **Standard** | **CPU Optimized** | **Memory Optimized** ||
|| Calico | 128 | 8&nbsp;192 | 3 vCPU | 12 GB | 4 vCPUs, 16 GB RAM (s-c4-m16) | 8 vCPUs, 16 GB RAM (c-c8-m16) | 4 vCPUs, 32 GB RAM (m-c4-m32) ||
|| Calico | 128 | 14&nbsp;080 | 6 vCPU | 21 GB | 8 vCPUs, 32 GB RAM (s-c8-m32) | 16 vCPUs, 32 GB RAM (c-c16-m32) | 8 vCPUs, 64 GB RAM (m-c8-m64) ||
|| Calico | 256 | 16&nbsp;384 | 6 vCPU | 24 GB | 8 vCPUs, 32 GB RAM (s-c8-m32) | 16 vCPUs, 32 GB RAM (c-c16-m32) | 8 vCPUs, 64 GB RAM (m-c8-m64) ||
|| Calico | 256 | 28&nbsp;160 | 11 vCPU | 42 GB | 16 vCPUs, 64 GB RAM (s-c16-m64) | 32 vCPUs, 64 GB RAM (c-c32-m64) | 16 vCPUs, 128 GB RAM (m-c16-m128) ||
|| Calico | 512 | 32&nbsp;768 | 12 vCPU | 48 GB | 16 vCPUs, 64 GB RAM (s-c16-m64) | 32 vCPUs, 64 GB RAM (c-c32-m64) | 16 vCPUs, 128 GB RAM (m-c16-m128) ||
|| Cilium | 128 | 8&nbsp;192 | 4 vCPU | 16 GB | 4 vCPUs, 16 GB RAM (s-c4-m16) | 8 vCPUs, 16 GB RAM (c-c8-m16) | 4 vCPUs, 32 GB RAM (m-c4-m32) ||
|| Cilium | 128 | 14&nbsp;080 | 7 vCPU | 26 GB | 8 vCPUs, 32 GB RAM (s-c8-m32) | 16 vCPUs, 32 GB RAM (c-c16-m32) | 8 vCPUs, 64 GB RAM (m-c8-m64) ||
|| Cilium | 256 | 16&nbsp;384 | 8 vCPU | 30 GB | 8 vCPUs, 32 GB RAM (s-c8-m32) | 16 vCPUs, 32 GB RAM (c-c16-m32) | 8 vCPUs, 64 GB RAM (m-c8-m64) ||
|| Cilium | 256 | 28&nbsp;160 | 13 vCPU | 51 GB | 16 vCPUs, 64 GB RAM (s-c16-m64) | 32 vCPUs, 64 GB RAM (c-c32-m64) | 16 vCPUs, 128 GB RAM (m-c16-m128) ||
|| Cilium | 512 | 32&nbsp;768 | 15 vCPU | 59 GB | 16 vCPUs, 64 GB RAM (s-c16-m64) | 32 vCPUs, 64 GB RAM (c-c32-m64) | 16 vCPUs, 128 GB RAM (m-c16-m128) ||
|#