Deploy CARTO spatial apps directly in the Snowflake Data Cloud
At the 2023 Snowflake Summit, taking place in Las Vegas, Snowflake announced Snowpark Container Services, launching in private preview and the public preview for Snowflake Native Apps. CARTO is excited to participate as a launch partner for both of these offerings, furthering our mission to unlock native geospatial capabilities directly within the Snowflake Data Cloud.
Since 2020, CARTO has been collaborating with Snowflake to bring advanced spatial analytics functionalities to the Snowflake Data Cloud. We are thrilled to expand this collaboration to include support for Snowflake’s latest product innovations.
Snowpark Container Services can be used as part of a Snowflake Native App (in development) to enable Snowflake customers to install and run sophisticated apps entirely in their Snowflake account. For Snowflake consumers, this means they can install cutting-edge products like CARTO’s Location Intelligence platform and run them in their Snowflake without ever having to move their governed data or compromise its security.
Let’s take a look at an example use case of CARTO and Snowflake in action!
The Vehicle Routing Problem, also known as VRP, is an optimization challenge aimed to determine the most efficient routes for a fleet of vehicles to visit a specific set of locations while adhering to various constraints.
To tackle this intricate spatial challenge, CARTO offers a comprehensive fleet optimization toolkit that empowers developers to fully leverage the robust and flexible computing resources of Snowflake. This toolkit enables the creation of routing applications that enhance operational efficiency, competitiveness, and environmental sustainability for companies in many different verticals.
First, CARTO’s Analytics Toolbox can be installed as a Snowflake Native App from Snowflake Marketplace, so its entire suite of geospatial functions runs within the customer’s Snowflake environment. The Fleet Optimization solution—which utilizes Snowpark Container Services—runs using Snowflake compute and accesses the natively integrated geospatial functions available via the native app.
At the heart of the Fleet Optimization architecture lies a Docker container operating within Snowpark Container Services, serving as the execution environment for all fleet optimization processes.
For data storage, Snowflake employs input tables to hold information like locations (deliveries), constraints (e.g., time windows), and vehicles (including additional limitations like vehicle capacities). These input tables conform to a predefined data model, ensuring data integrity within the optimization engine.
Similarly, Snowflake utilizes output tables to store the solutions generated by the optimization engine. Each solution consists of a collection of routes, comprising nodes (point geometry with a sequential order within the route) and edges (line geometry connecting each pair of sequenced nodes). These output tables also adhere to a predefined data model, preserving integrity with the optimization engine.
The Fleet Optimization solution produces various outputs, which are then visualized through an interactive dashboard equipped with maps and widgets in CARTO Builder.
Ready to take your geospatial analysis to new heights? Talk to us!
