# Steamboat Custom Indexes

Steamboat is our high-performance backend solution for `getProgramAccounts` (gPA) queries. By creating custom database indexes tailored to your specific application's needs, Steamboat can dramatically improve performance, reducing gPA query times from seconds to milliseconds. Because it uses the standard JSON-RPC API, no changes are needed to your application code.

The screenshot below shows a test where filtered `getProgramAccounts` requests for the Stake program were reduced from \~1700ms to \~30ms.

<figure><img src="/files/a848d99wZ6SXnLCsn4gz" alt=""><figcaption></figcaption></figure>

#### How It Works

Instead of scanning the entire chain state for every request, Steamboat serves `getProgramAccounts` queries from a dedicated database. This database is populated with indexes that are optimized for your application's specific query patterns.

Enabling Steamboat requires us to deploy the Steamboat Geyser plugin on your dedicated node. This plugin feeds on-chain data into a separate indexer database. This setup requires a node with sufficient hardware resources to handle the additional computational load.

#### Index Configuration

To ensure the best performance, the database indexes must be tailored to your workload. We offer two ways to configure them:

* **Client-Provided Indexes:** You can provide us with a list of the specific `memcmp` or other filters your application uses for `getProgramAccounts`. We will then create a set of custom indexes designed precisely for those queries.
* **Traffic Analysis:** Alternatively, our team can analyze your application's RPC traffic to automatically identify your most common and performance-critical `getProgramAccounts` queries. We then build the optimal set of indexes based on this real-world usage.

It is important to note that these indexes are tuned for your existing traffic. If your query patterns change significantly over time, the indexes may need to be refreshed periodically to maintain optimal performance.

#### Limitation: SPL Token Program

Steamboat is designed for custom programs and does **not** support `getProgramAccounts` queries for the standard SPL Token Program (`TokenkegQfeZyiNwAJbNbGKPFXCWuBvf9Ss623VQ5DA`). Performance for SPL token queries is already highly optimized by the built-in indexes included in the standard Agave validator software that our nodes run.

#### How to Enable Steamboat

Steamboat is available for dedicated node customers. Because it requires a custom setup and index configuration, please contact our support team to discuss your application's needs and get started.


---

# Agent Instructions: 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:

```
GET https://docs.triton.one/project-yellowstone/steamboat-custom-indexes.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.