Announcing Honeycomb for Snowflake: Secure, High-Performance Location Data Tools
Following Snowflake’s announcement of expanded Snowflake Native Apps at Snowflake Summit 2024, we’re excited to announce Honeycomb for Snowflake.
Honeycomb for Snowflake combines Honeycomb’s cutting-edge location data visualization with Snowflake’s industry-leading scale, security, and geospatial features.
With Honeycomb for Snowflake, companies can deploy internal tools that let their teams use location data to optimize operations while keeping sensitive data secure.
Snowflake’s scale + Honeycomb’s responsiveness
Honeycomb Maps runs completely in the browser, using advanced technologies like WebAssembly (Wasm) and DuckDB to make working with location data fast & responsive. However, while modern browsers are perfect for on-the-fly computation on medium-sized data, data still needs to be prepared and stored somewhere before it gets to Honeycomb.
That is where Snowflake comes in - its highly optimized, massively parallel cloud architecture can process huge amounts of data efficiently. Some of the largest companies in the world use Snowflake to store and process their business data, including massive amounts of spatial data.
By bringing Honeycomb to Snowflake, companies can efficiently prepare and filter data in Snowflake’s cloud data warehouse and then seamlessly visualize and interact with the data in Honeycomb’s browser-based tool.
Bring your visualization tool to your data, not the other way around
Honeycomb for Snowflake is deployed as a lightweight container which runs inside a Snowflake account using Snowpark Container Services. This means that you can view location data from Snowflake tables with Honeycomb in situ, without transferring the data outside of the Snowflake secure environment. As everything remains in the Snowflake environment, access control, including mandatory SSO, is handled by Snowflake.
None of your data ever touches Honeycomb servers, and it is not technically possible for anyone at Honeycomb to view your data.
By keeping your data inside the secure and governed Snowflake security perimeter, Honeycomb greatly reduces risk associated with data transfers, access management, and deletion requirements. This is particularly useful for location data which may be considered PII and have related obligations under GDPR and CCPA regulations.
Utilize Snowflake's innovative features
Honeycomb for Snowflake is purpose-built to fully utilize the capabilities of the Snowflake platform.
Snowflake Native H3 functions
Snowflake has been steadily increasing native support for GEOGRAPHY and GEOMETRY types and associated functions. In February Snowflake extended this functionality by adding native H3 functions, allowing for easy conversion between GEOGRAPHY objects and H3 indexes.
Honeycomb for Snowflake perfectly compliments this functionality, letting companies seamlessly go from H3 indexes in a table to H3 hexagons on a map.
For example, a Snowflake data table with billions of lat/lon points could be efficiently aggregated to H3 cells using Snowflake’s H3_POINT_TO_CELL() function, before being interactively visualized using Honeycomb.
Efficient data transfer with Apache Arrow
In addition, both Snowflake and Honeycomb natively support the next-generation columnar Arrow IPC data exchange format.
This means that data from Snowflake is streamed to the Honeycomb client without any serialization or deserialization costs (and then Honeycomb sends this same data to the GPU for rendering using GeoArrow).
This enables buttery-smooth visualization of previously overwhelming amounts of geospatial data.
What's next: building for key use cases
We are building the first version of Honeycomb for Snowflake around two key use cases in the mobility and logistics industry:
H3 market overview maps
- City managers at a bike rental company should be able to see the real-time supply (bike locations) and demand (app open events) maps at H3 resolution 10 level.
Zone management with write-back
- With the H3 supply/demand map from above, a city manager at the bike-sharing company should be able to edit the boundaries of a surge pricing zone and save the new polygon back to a Snowflake table.
Honeycomb in Snowflake is currently in private preview and available to select customers. If this sounds interesting, please reach out. We hope to make it publicly available on the Snowflake Marketplace in the near future.
-Carston
This blog post includes references to Snowflake, a registered trademark of Snowflake, Inc. Maps and Data LLC is not affiliated with, endorsed by, or sponsored by Snowflake, Inc. The mention of Snowflake is for informational purposes only and does not imply any association with or endorsement by Snowflake, Inc.