Snowflake Pledges Row-based Storage Engine For Transactional Data

Cloud data warehouse specialist Snowflake is broadening its toolset to allow devs to build applications inside its platform, while providing a new row-based storage engine to support analytics on transactional data.

Launched at its annual conference this week, the features are part of a plan to encourage users – and investors – to no longer see it as a mere cloud data warehouse and to view it more as a platform for sharing data and data-analytics applications.

Snowflake is supporting new transactional workloads in something it calls Unistore, which is based on Hybrid Tables supported by a new row-based storage engine to better handle transactional data.

Users can either build data pipelines to stream data from transactional systems into Snowflake, or build transactional applications themselves within the platform.

The goal will be to create “a data pipeline to pull all that data into Snowflake [and then] everything's in Snowflake and it's easy to manage. [It] reduces or entirely removes the complexity of managing your transactional data,” said Carl Perry, director of product management at Snowflake, on a media call.

Another approach might be to perform analytics on data in the transactional database system itself, as vendors such as SAP, MongoDB and Oracle claim to support.

Meanwhile, Snowflake — which was once valued at $120 billion following its 2020 IPO — is set to introduce a development environment to help devs build applications and even sell or share them on its platform.

Native Application Framework — which like Unistore is only available in private preview — promises to help developers build applications using Snowflake functionalities such as stored procedures, user-defined functions (UDFs), and user-defined table functions (UDTFs).

Snowflake offers Streamlit integration for developing interactive customer interfaces and telemetry features including events and alerts for monitoring.

'Real-time' use cases a 'big step forward'

Hyoun Park, CEO and chief analyst at Amalgam Insights, said Snowflake customers saw transactional data as a "gap" in Snowflake's offering and Unistore was a big step forward. "Snowflake's claim to fame has been around its powerful columnar analytic capabilities that made it a scalable data warehouse to analyze historical data, but attaching transactional data took extra steps to insert transactional data from another source.

"Transactional data is the key to what is commonly thought of as 'real-time' use cases used for eCommerce, retail, personalization, financial services, and logistics management and Snowflake now has a native solution to ingest transactions and then process the data analytically. This is a big step forward for Snowflake to be a fully functional data cloud as Unistore fills a big gap that customers have asked for," he said.

The success of Snowflake's application development environment would depend on how devs take to it, Park said.

“Snowflake is following a long-worn path first created by Oracle and other major database providers. Now that Snowflake has a data gravity from the sheer volume of data that it supports, it has the foundation to be an application platform as well.

"It's always easier to build the app directly on top of the data if the application development environment is relatively easy to use. Snowflake's Native Application Framework is still in Private Preview, so it remains to be seen what the uptake is in developing apps on the Snowflake Marketplace,” Park said.

• Snowflake stock drops as some top customers cut usage
• Real-time data analytics firm Tinybird nets $37m in Series A
• Snowflake rolls back growth estimates, wipes 30% off value
• 'Enterprise' browser maker Island valued at $1.3bn out of the gate

Snowflake's expansion from data warehousing to a data cloud was a response to the "near impossible" expectations that Wall Street has placed on it, he added.

Along with other tech vendors, Snowflake has seen its value crash in recent weeks, dropping to $36 billion. But this still put it on a par with much larger enterprise vendors such as Dell, HP, and Workday. "Snowflake's expectations from institutional investors are that it will be a dominant enterprise data source and Snowflake needs to build out all of the key areas that monetize data to fully realize those expectations," Park said.

Snowflake launched new security features and expanded its support for data science language Python — first announced last year — at its annual conference. ®