## Sign Up

[Sign-up](🔗) for a Rockset account using your GitHub account, Google account, or email address. During the course of your two-week trial, you will have $300 worth of credits to use however you see fit. No credit card required! Billing is determined by your compute (Virtual Instance size) and storage usage. After your trial ends, you can continue to use Rockset for free by using the FREE Virtual Instance and staying below the 2GB of free storage.

**Pro-tip:** If you’re using a larger VI for your trial, switch it back to Shared VI when you’re not actively using it to save credit! For more information on Billing, click [here](🔗).

## Create a Collection

Now that you've created an account and understand how to use your $300 free trial credits, let's navigate to the [Collections tab of the Rockset Console](🔗) where we will create your first collection!

A [collection](🔗) in Rockset is a set of Rockset documents. Similar to tables in traditional SQL databases, collections can be queried using SQL, either directly or using [Query Lambdas](🔗).

In this tutorial, we will create two collections from public datasets hosted on AWS S3:

• **The Movies Dataset: Film Releases**: A sample of movies and information including their genre, popularity, and revenue.

• **The Movies Dataset: Film Ratings**: A sample of movie ratings by user.

Both datasets are publicly available [here](🔗).

### Create the Film Releases Collection

Follow the steps below to create the Film Releases collection:

1. Click **Create your first Collection** in the [Collections tab of the Rockset Console](🔗):

   

   

2. Select **Public Datasets** as the data source for your collection:

   

   

3. Select the dataset **The Movies Dataset: Film Releases** and click **Start** below:

   

   

4. Configure and create the collection:

   • Under the **Transform Data** section, you will see that an [Ingest Transformation](🔗) has already been predefined for you. This transformation simply processes incoming data as it is written into your Rockset collection by cleaning up the data and defining data types for a few fields. For the purposes of this tutorial, we recommend that you do not change the predefined ingest transformation as our queries later on may be affected.

   • Name your collection `film_releases` in the **Collection Name** field.

   • When you're ready, click **Create** at the top right of the page to complete the creation of your collection.

The source preview is automatically generated so you can explore the semi-structured JSON data in a tabulated form:

The collection creation process will take about 3 minutes to complete. Its initial status will be **Created**, after which, it will change to **Ready**. At this point, documents will begin to flow into the collection gradually until the entire dataset is ingested. When completed, you should see something like this:

**Note:** You may need to refresh the screen for the status to update.

### Create the Film Ratings Collection

Now repeat the same steps above to create the **Film Ratings** collection. This time select the dataset **The Movies Dataset: Film Ratings** and name your collection `film_ratings` in the **Collection Name** field.

## Execute a Query

Now that both collections have been set up, we can use SQL to query the two collections.

### Sample Query

Below is a sample query to suggest movies to a user based on their genre preference and the movie's rating. Since `genre` is an array field (as a single movie may fit multiple genres), we use `UNNEST` to expand this array and create a record for each `(genre, movie)` pair. We also exclude movies rated by a specified user.

Follow the steps below to enter and use the query:

1. Navigate to the [Query Editor tab of the Rockset Console](🔗) to start writing and executing SQL queries.

2. Copy the query below into the SQL editing area:

   Text

   

3. Click **Run** to execute the query. The **Results** tab below the query shows the rows returned from the query:

   

   

### Sample Query with Parameters

In the above query, we used the `Action` genre and user `100`. Now, let's make these values parameters that can be specified at runtime. Follow the steps below to add and test these parameters:

1. Click the **+** next to **Parameters** located below the query name tab to create a new parameter:

   

   

2. Populate the parameter details with the following and click return on your keyboard to create the new parameter:

   • Set **Parameter Name** to `genre`.

   • Set **Type** to `string`.

   • Set **Parameter Value** to `Action`.

   

   

3. Repeat Step 2 with the following parameter details and click return on your keyboard to create the new parameter:

   • Set **Parameter name** to `user_id`.

   • Set **Type** to `int`.

   • Set **Parameter Value** to `100`.

4. Modify the SQL statement from the previous section to incorporate the parameters created in steps 2 and 3 above:

   • Replace `genres.name = 'Action'` with `genres.name = :genre`.

   • Replace `r.user_id = 100` with `r.user_id = :user_id`.

Here is the new SQL statement with these updates:

6. Click **Run** to execute the query. The **Results** tab below the query shows the rows returned for the `Action` genre with the User ID `100`:

   

   

## Next Steps

This completes the quickstart tutorial! Here are some suggestions for next steps:

• Keep building by checking out some of the pages below to continue exploring Rockset:

  • [Connect your external data source](🔗)

  • [Explore the SQL Reference](🔗)

  • [Explore the REST API Reference](🔗)

  • [Check out our Developer Tools](🔗)

• Invite other members of your team to your organization using their email address in the [Users tab of the Rockset Console](🔗). You can determine if each new user should have Administrator, Member or Read-Only access to Rockset. Learn more about user management [here](🔗).

• Join us in our [Rockset community](🔗) and share what you are looking to build with Rockset. We’re hanging out and ready to answer your questions!