> ## Documentation Index
> Fetch the complete documentation index at: https://langchain-5e9cc07a-preview-docsse-1788217470-b7511b1.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Configure SmithDB for scale

> Choose tested SmithDB resource tiers and size Kubernetes workloads and the PostgreSQL metastore.

<Note>
  SmithDB is optional and available in beta. Use these tiers only after you decide to enable SmithDB. For non-SmithDB LangSmith services, see [Configure LangSmith for scale](/langsmith/self-host-scale).
</Note>

Use these tested tiers as starting configurations for SmithDB.

<Note>
  Configure observability before tuning capacity. See [Configure SmithDB observability](/langsmith/self-host-smithdb-observability).
</Note>

## Choose and scale a tier

<Note>
  These tiers are starting points, not fixed throughput limits or prescribed configurations. CPU, memory, and ephemeral-storage values are per replica.
</Note>

Choose the tier whose tested ingestion and query rates most closely match or exceed your expected steady-state load.

Apply the per-replica resources and starting replica counts from the selected tier. If the workload later outgrows that starting point, SmithDB compute can be scaled horizontally.

<Note>
  SmithDB query, ingestion, and compaction-worker HPAs are enabled by default. They require Kubernetes Metrics Server or another `metrics.k8s.io` provider. Verify availability with `kubectl get --raw /apis/metrics.k8s.io/v1beta1`.
</Note>

## Baseline tiers

| Component         | Small<br />10 ingest / 10 query QPS                          | Medium<br />100 ingest / 40 query QPS                          | Large<br />1000 ingest / 100 query QPS                           |
| ----------------- | ------------------------------------------------------------ | -------------------------------------------------------------- | ---------------------------------------------------------------- |
| Ingestion         | 1 replica<br />4 CPU<br />8 Gi memory<br />100 Gi ephemeral  | 1 replica<br />16 CPU<br />32 Gi memory<br />100 Gi ephemeral  | 2 replicas<br />56 CPU<br />150 Gi memory<br />1000 Gi ephemeral |
| Query             | 1 replica<br />4 CPU<br />8 Gi memory<br />200 Gi ephemeral  | 1 replica<br />28 CPU<br />48 Gi memory<br />200 Gi ephemeral  | 4 replicas<br />28 CPU<br />50 Gi memory<br />1000 Gi ephemeral  |
| Compaction        | 1 replica<br />2 CPU<br />4 Gi memory                        | 1 replica<br />4 CPU<br />8 Gi memory                          | 1 replica<br />8 CPU<br />16 Gi memory                           |
| Compaction worker | 1 replica<br />8 CPU<br />16 Gi memory<br />100 Gi ephemeral | 3 replicas<br />16 CPU<br />32 Gi memory<br />100 Gi ephemeral | 4 replicas<br />28 CPU<br />50 Gi memory<br />300 Gi ephemeral   |
| Cluster manager   | 1 replica<br />250m CPU<br />256 Mi memory                   | 1 replica<br />250m CPU<br />256 Mi memory                     | 1 replica<br />2 CPU<br />2 Gi memory                            |

## Configure resources with Helm

The chart defaults to `small`. Select `small`, `medium`, or `large`:

```yaml theme={null}
smithdb:
  resourceTier: medium
```

A tier sets per-replica CPU, memory, and ephemeral-storage requests and limits. Replica counts and autoscaling are configured separately. The chart enables autoscaling for query, ingestion, and compaction worker by default.

For local-disk components, the ephemeral-storage limit also sizes the generated `emptyDir`. Explicit component `resources` or `volumes` settings take precedence.

See [SmithDB resource tiers](https://github.com/langchain-ai/helm/tree/main/charts/langsmith#smithdb-resource-tiers) for current values and chart details.

<Accordion title="Example explicit resource configuration">
  To configure resources without `smithdb.resourceTier`, use the component-level values directly. This medium-tier example is also a copyable reference for overriding individual components:

  ```yaml theme={null}
  smithdb:
    query:
      deployment:
        resources:
          requests:
            cpu: "28"
            memory: "48Gi"
            ephemeral-storage: "200Gi"
          limits:
            cpu: "28"
            memory: "48Gi"
            ephemeral-storage: "200Gi"

    ingestion:
      deployment:
        resources:
          requests:
            cpu: "16"
            memory: "32Gi"
            ephemeral-storage: "100Gi"
          limits:
            cpu: "16"
            memory: "32Gi"
            ephemeral-storage: "100Gi"

    compaction:
      deployment:
        resources:
          requests:
            cpu: "4"
            memory: "8Gi"
          limits:
            cpu: "4"
            memory: "8Gi"

    compactionWorker:
      deployment:
        resources:
          requests:
            cpu: "16"
            memory: "32Gi"
            ephemeral-storage: "100Gi"
          limits:
            cpu: "16"
            memory: "32Gi"
            ephemeral-storage: "100Gi"

    clusterManager:
      deployment:
        resources:
          requests:
            cpu: "250m"
            memory: "256Mi"
          limits:
            cpu: "250m"
            memory: "256Mi"
  ```

  When configuring local-disk components manually, keep `limits.ephemeral-storage` and any explicit `emptyDir.sizeLimit` consistent with your Kubernetes policy.
</Accordion>

## Metastore capacity

The baseline tiers cover SmithDB Kubernetes workloads only. They do not include the PostgreSQL metastore. Use these provisional starting points for a dedicated metastore:

| Tier   | vCPU | Memory |
| ------ | ---- | ------ |
| Small  | 2    | 16 GiB |
| Medium | 4    | 32 GiB |
| Large  | 8    | 64 GiB |

Choose the nearest supported PostgreSQL instance shape from your provider and monitor database resource use and transaction latency during rollout.

## See also

* [Install LangSmith with SmithDB](/langsmith/self-host-smithdb-install)
* [Prepare SmithDB supporting infrastructure](/langsmith/self-host-smithdb-infrastructure)
* [Configure SmithDB observability](/langsmith/self-host-smithdb-observability)
* [Configure LangSmith for scale](/langsmith/self-host-scale)

***

<div className="source-links">
  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>

  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/langsmith/self-host-smithdb-scale.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>
</div>