kdb Insights Enterprise
The Data Timehouse
kdb+ Time Series Database
PyKX Python Interoperability
Services & Support
Industry & IoT
Energy & Utilities
Healthcare & Life Sciences
KX University Partnerships
Partner with Us
Become a Partner
Find a Partner
Connect with Us
KX and Snowflake are jointly collaborating on a strategic partnership to extend KX’s industry leading Data Timehouse technology to Snowflake customers.
The power to analyse and gain insights from temporal (time) and vector (space) data is the driving force behind digital transformation and generative AI, and there is growing demand from Snowflake customers to run these workloads on the Snowflake cloud. Our partnership addresses that demand by enabling data scientists, data engineers and developers to run complex machine learning and AI queries using kdb inside their Snowflake environment, delivering significantly higher levels of performance while reducing complexity and cost.
Look further than BI. Bring supersize machine and deep learning model training, search, optimization and time-series models to Snowflake with KX.
Execute production-grade analytics from Python code and Jupyter Notebooks with 80X more performance, right inside Snowflake.
Combine raw data with vector embeddings. Generate, store, and run similarity search on vector embeddings, absolutely GPU-free!
Easily integrate warehouse analytics with feature stores and transition to real-time machine learning inference models at speed.
Don’t force-fit time and vectors into conventional data stores. Time is the ultimate “natural sequence” needed to find patterns and meaning. AI & ML need time and vector data. With KX, bring native time series and vector processing to Snowflake schema.
Go beyond simple SQL queries. Run full production analytics right inside your Snowflake Cloud Data Warehouse, with greater cost predictability from the most efficient compute in the world.
By integrating streaming, embeddings generation, VectorDB, raw data, timeseries, and analytics directly into the data warehouse and implementing fully integrated real-time and historical analytics outside it, reduce and even eliminate unnecessary lakehouse complexity.
Python is the name of the Data Science game as Snowflake and Snowpark have realized. By eliminating distance between Jupyter Notebooks production-grade models, the data warehouse and SQL, bring data scientists, data engineers and developers together to improve product lifecycles, including real-time MLOps pipelines.
With Snowflake and KX, power generative AI. Unlike other so-called vector databases, run ultra-efficient GPU-free CPU-based similarity search on vector embeddings, lowering costs and increasing accessibility and get access to raw stored data.
EXAMPLE KDB SNOWFLAKE USE CASE
For an e-commerce, payments or financial institution this real-time payments analytics process shows how you can inform model development directly in the Snowflake cloud data warehouse through the Python enabled interface. In this case, we directly connect live inference back to the warehouse as well as take live inference from Snowflake-initiated feature extraction.
Developers can benefit from KX for Snowflake, a Python-first integration of kdb Insights that enables users to run supersized Machine Learning (ML) and AI workloads and vector-processing for model development and training, and vector search for generative AI. It also enables the development of time-series models for use cases including anomaly detection, risk strategies, predictive healthcare and maintenance.
KX for Snowflake will be available initially via an Anaconda distribution on Snowpark, Snowflake’s developer framework that brings support for Python, SQL, and other programming languages to the Snowflake environment.
Run SQL and Business Intelligence queries as normal, but access considerably more powerful models including generative AI.
Simplify your data storage, retaining the integrity of your Snowflake governance with practical business needs of analytics, AI and compute requirements.
Enable data scientists to run production-grade model execution in Snowflake directly from their Jupyter Notebooks and Python code.
Integrate in-warehouse analytics, for example model training, with real-time machine learning inference model at the edge.