If you would like your announcement to be included in Amalgam Insights’ weekly data and analytics roundups, please email email@example.com.
Snowflakes in June
The biggest news: Snowflake Summit 2022 was this week, and a wide variety of data companies released announcements in conjunction with the conference, whether technical or fiscal in nature.
Snowflake Releases Unistore, A Workload Combining Transactional and Analytical Data in One Platform
Snowflake itself had several major announcements at Snowflake Summit 2022. The first covered the debut of Unistore, a workload that will allow Snowflake users to store transactional and analytical data together in a Snowflake data warehouse. Snowflake’s new Hybrid Tables will enable this new approach; customers will be able to perform fast analytics on transactional data stored in Snowflake for more timely understanding, and build transactional apps atop Snowflake.
Snowflake Introduces Native Application Framework
Snowflake also announced a Native Application Framework. Developers will be able to build data applications on Snowflake and monetize them on the Snowflake Marketplace, allowing Snowflake consumers to install and run those applications securely in their own Snowflake instances without needing to move or share data. In conjunction with this, Informatica launched their new enterprise data integrator on Snowflake, reflecting an expanded partnership with Snowflake.
The Register interviewed Amalgam Insights’ Hyoun Park on Snowflake’s Announcements, covering Unistore and the Snowflake Native Application Framework.
Snowflake Expands Native Python Support and Data Access with Snowpark for Python
Finally, Snowflake announced a number of changes demonstrating stronger Python support for machine learning and application development on Snowflake. First, Snowflake launched Snowpark for Python into public preview, broadening from existing Scala and Java support. This means that Python’s open-source packages and libraries are now accessible within Snowpark, providing a strong foundation for the most popular language for building machine learning models. Additional support for Python developers includes a new Streamlit integration for easier app development on Snowflake; Snowflake Worksheets for Python to enable development of machine learning models, pipelines, and applications directly in Snowsight; large memory warehouses to support memory-intensive operations like feature engineering and model training on large datasets, enabled through Snowflake’s Anaconda integration; and SQL Machine Learning, allowing data analysts to more easily use machine learning algorithms without requiring advanced knowledge. The first algorithm available is time-series forecasting. Finally, Snowflake also increased data access with better support for ingesting and transforming streaming data, and working with data external to the Snowflake Data Cloud, even on-prem data, while still conferring some of the advantages of storing data in Snowflake.
Domino Data Lab Reveals Investment from Snowflake Ventures
Domino Data Lab announced an investment from Snowflake Ventures this week for an undisclosed amount, following up on Snowflake Ventures’ previously unannounced participation in Domino’s Series F funding round last October. The additional investment demonstrates the strength of the Snowflake-Domino partnership being robust enough for Snowflake to take an equity stake in Domino, rather than being solely a technical partnership.
Matillion Announces Snowflake Ventures Investment
Data integration platform Matillion also announced an investment from Snowflake Ventures. As part of this ongoing “investipartnership,” Matillion will be among the first Snowflake data integration partners to use the just-announced Snowflake Native Application Framework by making Matillion connectors available directly within Snowflake.
Launches and Updates
KNIME Software Release: Improved Python, Snowflake Integrations
KNIME announced the latest release of their data science platform. Key new features include upgrades to KNIME’s Python support with a built-in Python environment and the ability to write KNIME extensions entirely in Python, as well as a Snowflake integration that allows users to build machine learning models in H2O.ai, and then push the model down to Snowflake for predictions.
Okera Now Generally Available on Snowflake
Data security and governance company Okera announced that Okera was now available on Snowflake as a SaaS offering for Snowflake Data Cloud. Okera’s universal data authorizaton policies are automatically translated into Snowflake data access governance controls, allowing native data security policy enforcement within Snowflake.
Yellowbrick Launches Latest Version of its Data Warehouse
Cloud data warehouse Yellowbrick released a new version of its platform this week. Key features include on-prem and AWS deployment options (Azure and Google Cloud Platform coming in Q3), data lake integration using Parquet, separation of compute and storage for more elastic scaling on demand, and multiple payment models (consumable either on-demand or through a subscription based on fixed capacity). Yellowbrick also announced two new partnerships with Saarthee and Saxon, two data and analytics companies.
H2O.ai Expands Snowflake Partnership
H2O.ai continues to grow its Snowflake partnership. Users are able to use H2O.ai machine learning capabilities on the data within their Snowflake environment; H2O.ai is expanding support for financial services, manufacturing, and healthcare customers doing machine learning.