Top data innovations announced at Snowflake Summit 2022

We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 – 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!

As enterprises continue to grow bullish on data, Snowflake is moving the needle on its data cloud offering to cover all emerging bases and provide organizations with a comprehensive solution to generate insights and build powerful apps as quickly and efficiently as possible.

At its annual conference in Las Vegas, the Montana-headquartered company has announced major data innovations to enhance the core capabilities of its product, as well as unlock new opportunities for enterprises and help them support advanced use cases. The focus, as Snowflake’s senior vice president of product, Christian Kleinerman emphasized, continues to be on three key areas: Improving the overall performance of the platform, strengthening governance and extending the reach and types of data it can cover.

“Our latest innovations extend the value of our customers’ data-driven ecosystems, enabling them with more access to data and new ways to develop with it directly in Snowflake,” he said. “These capabilities, paired with Snowflake’s best-of-class data security and privacy, are changing the way teams experiment, iterate, and collaborate with data to drive value.”

Below is a rundown of the major announcements.

Snowflake Unistore

First, Snowflake announced the launch of a new workload for its data cloud – Unistore. Available in private preview, the solution extends the capabilities of the platform and provides enterprises a way to do away from the practice of storing traditional and analytical data across different, siloed systems.

Organizations usually keep their analytical and transactional datasets separate. The practice gets results, but also increases governance complexity and affects the speed of action. After all, the data has to be moved from systems occasionally. Unistore solves this problem by enabling organizations to leverage and manage the lifecycle of both transactional and analytical datasets in a single location. 

The workload includes hybrid tables, which provide fast single-row operations and enable enterprises to build transactional business applications directly on Snowflake. Teams can even perform swift analytics on hybrid tables to get insights from their transactional data or join them with other existing Snowflake tables for a holistic view of all data.

Unistore is currently available to select customers, including Adobe, UiPath, Novartis, IQVIA and Wolt.

Native Application Framework

The next big thing coming from Snowflake’s summit is the announcement of Native Application Framework to help enterprises build, distribute and deploy applications natively in the Data Cloud. 

Currently in private preview, the solution provides developers access to Snowflake functionalities, like stored procedures and user-defined functions, as well as third-party integrations and telemetry features to build and troubleshoot applications with interactive customer interfaces.

Once an app is developed, it also allows users to monetize it on the Snowflake Marketplace, making it instantly available for use within the Snowflake instance of over 6,000 companies. In fact, LiveRamp, Capital One Software and Informatica have already used the framework to develop applications for use cases such as identity resolution, cloud cost management and data integration.

Snowpark for Python

Snowflake’s developer framework, Snowpark, is getting Python support. According to the company, the move will make Python’s rich ecosystem of open-source packages and libraries seamlessly accessible in the Snowflake Data Cloud, further improving the programming environment it provides to build scalable pipelines, applications, and machine learning (ML) workflows.

Snowpark for Python, currently in public preview, comes with a highly secure Python sandbox and runs on the same compute infrastructure as Snowflake pipelines and applications written in other languages.

“At its core, Snowpark is all about extensibility, and Snowpark for Python provides us with the tools we need to work with data effectively in our programming language of choice,” Joe Nolte, AI and MDM architect at the Allegis Group, said. “Snowpark is becoming our preferred framework for data science and application development, providing our teams with a seamless experience to easily collaborate with data and bring everyone onto the same platform for accelerated time-to-value.” 

Improved data access

To improve enterprise users’ access to data, Snowflake also introduced several innovations, including Snowpipe Streaming. The feature, as the company explains, will enable serverless ingestion of streaming data and is set to be complemented with Materialized Tables, designed to simplify the transformation of streaming data.

In addition, the company announced Iceberg Tables to enable users to work with Apache Iceberg, a popular open table format, in external storage and the ability to manage, classify, tag and govern on-prem data. The latter will be useful in cases when data has not been moved to the cloud or is yet to be moved.

Other notable improvements

Among other things, Snowflake improved geospatial support on its platform with a new geometry data type that provides enterprises with a planar coordinate system. It also expanded replication capabilities on the platform with account replication, covering all sorts of metadata about account users, as well as pipeline replication, where Snowflake will automatically do a fail-over to make sure there are no gaps or duplicates in the data being loaded. 

On the governance side, the tagging and masking features of the platform have been extended to offer tag-based masking. As part of this, Snowflake will be able to automatically redact or partially redact data depending on the annotations and the user querying the data. The company also announced column-level lineage tracking and a new user experience where enterprise users could easily see their data governance policies.

Originally appeared on: TheSpuzz