API overview

The API is an interface that allows users to interact programmatically with the Matillion Data Productivity Cloud.

The API follows a RESTful architecture, which is characterized by several key attributes:

A resource based URL format.
When interacting with the API use form encoded data.
Response from the API is in JSON format.
The Matillion API uses HTTP response codes.
The API employs standard authentication methods, including the use of bearer tokens, to secure access. It also follows standard HTTP verbs (e.g. GET, POST, PATCH, and DELETE) for consistent and familiar interaction with the API.

Before you begin

Create a Hub account or sign in.

Base URL

EU server:

https://eu1.api.matillion.com/dpc

US server:

https://us1.api.matillion.com/dpc

Note

Make sure to choose the right endpoint when you are using the console. Your Hub account needs to be in the same region as the endpoint server.

Authentication

The Data Productivity Cloud API supports Bearer token authentication. To authenticate with the API endpoint, you should provide a token obtained through API credentials created in your Hub account.

For detailed instructions on the authentication process, read authentication.

Rate limits

The API has a fixed rate limit, which may cause an HTTP error if you attempt a large number of concurrent endpoint calls.

The HTTP error code returned if the rate limit is exceeded is 429 Too Many Requests. The Retry-After HTTP response header will indicate how long you need to wait before making a subsequent request.

The API rate limit is configured as follows:

Average: 60 requests per minute.
Burst: 5 requests per second.

The average limit of 60 requests per minute equates to 1 request per second; however, the burst rate lets you exceed this per-second rate, bursting to a maximum of 5 requests per second, while still not allowing a total of more than 60 requests per minute.

If you receive a rate limit error from the API, try reconfiguring your scripts or pipelines to make use of sequential endpoint calls. In a Data Productivity Cloud pipeline calling the API, consider including a Retry component with the Retry delay property set to an appropriate value. We recommend a setting of Long delay with backoff, with at least five retries.