Getting started with Sledhouse

This article provides you with a step-by-step on how to get started with Bobsled Sledhouse.

Prerequisites

1. Bobsled Account: You must have access to Bobsled’s main application and the Sledhouse tab/feature enabled in your environment

Step 1: Create a Sledhouse Table

A Sledhouse Table is the internally replicated version of a source table. Once set up, Sledhouse keeps it in sync according to your chosen replication schedule and pattern (append, append and update, etc.).

1. In the Bobsled Application, locate and select the Sledhouse Tables Tab, then select the “Create Sledhouse Table” button

2. In the wizard, select the data source you have configured before (e.g., BigQuery)

3. Choose a table from the list and click “continue”

4. Choose a name for your Sledhouse Table (a human-friendly name) and an Alias (A short, SQL-safe identifier when referencing this table in data products later)

5. Optionally review your schema and confirm the column structures following Bobsled data types

6. Set a replication pattern for your Sledhouse Table:

   • Append: Every new record in the source is appended

   • Update & Append: Provide a unique identifier and a “last updated” timestamp to merge changes

   • Overwrite: Replace the table each time with a full snapshot

   
   

TIP:
The available replication patterns work in the same was as in Cloud Data Warehouse Sources to Cloud Data Warehouse Destination in Bobsled transfers.

7. Optionally set Partitioning for the Sledhouse Table.

   • Choose one or more columns to partition for performance optimization. If you anticipate frequent queries filtered by a specific column (e.g., state, zip_code, or event_date), use it as a partition key to speed up downstream queries.

TIP:
• Large datasets will significantly benefit from partitioning.
• When using time-based data types, Bobsled offers transformation options to tune the right field cardinality by choosing the right timeframe to partition by (hour, day, month, year).

NOTE: Clustering/zOrder support (in addition to partitioning) is coming soon to give further performance tuning options.

8. Set the scheduling options.

   • Currently, Sledhouse runs a default hourly sync

   • Bobsled offers two ways of setting syncing preferences:

     • Simple scheduler (a set of fixed intervals up to 1 month cadence) and,

     • Cron scheulder (an advanced scheduling tool to suit your operational needs).

NOTE:
• Sledhouse defaults to hourly syncs and does not support intervals shorter than one hour. If you require more frequent syncs, please contact your account team.

• Bobsled Cron scheduler uses UNIX cron syntax ↗

9. Once happy with your Sledhouse Table configuration, click “Continue“

10. Review and optionally publish your Sledhouse Table as a Data Product

    • This immediately makes the table available for sharing as-is

11. Click “Save”

Managing Sledhouse Tables

On the Sledhouse Table page, you can review the state of the pipeline, and also:

• Schema: Displays all columns replicated from the source, eventually enabling type casting or other transformations

• Logs: Shows ongoing and past replication jobs, with status and potential errors

• Settings: This lets you revisit the configuration (e.g., replication pattern, partitioning) and make changes

Step 2: Create a Data Product

A Data Product is a curated subset (or transformation) of your Sledhouse Table or tables. It can filter rows, select a subset of columns, or join multiple Sledhouse Tables together. Data Products are what you ultimately share with consumers via Bobsled’s standard sharing mechanism. By default, Data Products function as views in the destination data warehouse. Future releases may allow more advanced materialization.

1. In the Bobsled Application, locate and select the Data Products Tab, then select the “Define Data Product” button

2. In the wizard, select the Sledhouse Table you’d like to define a Data Product on and click “continue”

3. Choose a name for your Data Product (a human-friendly name) and an Alias (A short, SQL-safe identifier). On this page, Bobsled offers two ways to define Data Products:

Option 1: Via interactive mode (no code)

In Interactive mode, a user can choose to include or exclude certain columns from a sledhouse Table in a data product and use a no-code query build to control the rows returned in the data product. This enables fulfillment teams to control data entitlements without writing any code.

• In the “Interactive mode” tab a user chooses two configuration options: columns in the data product and applying row filters

1. Column selection

Column selection provides the following options for selecting which columns are available in a given data product.

• Included / Excluded Columns:

  • Select which columns should be included or excluded from the data product. By default all columns start as included and the user must select which columns should be excluded. The buttons in the middle allow for columns to move quickly between sections. Columns are also able to be dragged and dropped between the two squares.

• Schema evolution:

  • Turning on schema evolution allows for automatically flowing new columns in a Sledhouse Table down to the data product. This allows for simpler management for data products where schemas are often evolving and new columns need to pass through to customers.
    

2. Row filtering:

   • Row filtering allows for using an easy-to-use query builder that allows for defining filters (predicates in an SQL statement) that control what data is returned. This feature will let you quickly define filters like “event_date is in the past 30 days”.

     Multiple filters can be combined using AND or OR semantics to create almost any data product definition required. For more complex data products the SQL builder and co-pilot (See Option 2) can be used.

     

Option 2: Via SQL Editor

1. In the “SQL editor” tab, a built-in SQL editor lets you write transformations using Snowflake-compatible syntax.

   • Reference your Sledhouse Tables by their aliases using {{ sht('alias') }} syntax.

   • Example:  

     sql
     Copy
     SELECT *
     FROM {{sht('sht_alias'}}
     WHERE state = 'CA'

2. Optionally use the AI-Assistand (Copilot), you can type a natural-language request, and Sledhouse will suggest SQL statements. You can edit or refine these suggestions as needed

3. Once happy with your SQL, and no errors are observed, click “Save“, and the Data Product is ready to share

Managing Data Products

On the Data Product page:

• Settings: This lets you revisit the configuration (e.g., the SQL and source Sledhouse Table) and make changes

• Schema: Displays all columns set for this Data Product

NOTE:
Schema for SQL-generated Data Products is not available yet.

Step 3: Sharing Data Products

Data Products are the shareable units in Sledhouse. Once a Data Product is defined, you can share it with consumers via Bobsled’s standard share interface. The primary “destination” is Snowflake shares (with more destinations to come).

1. In the Bobsled Application, locate and select the Shares Tab, then select the “Create Share” button

2. In the Share page select Source and choose Data products

3. Select and choose your Destination and region and enter the required access credentials

4. Once finished, click “Share Data Products’ on the Share page

5. Choose your Data Products and click Share.

You’ve now successfully shared Data Products in Sledhouse Public Preview. We look forward to your feedback and to helping you integrate Sledhouse into your data architecture.

Schema Evolution

• Sledhouse Table Schema: Automatic updates of the Sledhouse Table schema and downstream data products are coming soon.

• Data Product Schema: If the underlying table schema changes, your Data Product may need to be updated accordingly—especially if you’re selecting columns or applying filters on changed fields.