Snowflake has released news at its annual Conference of Torrent Users today, from the new adaptive ability to calculate and integrate open flow data to new analyticity and AI enhancements such as AI SQL and new AI agents. The company also announced its intention to obtain crunchy data, a postal database and services provider.
Snowflake and 16,000 of its customers were taken over by the Moscone Center in San Francisco to host the summit. Snowflake, formerly called Cloud Summit data, used its show as a launchpad for several announcements, including its intention to get crispy data that operates the hosted version of Postgres and develops a postress engine that works with parquet data. For more information about crunchy, check out the article we wrote about them last July.
Meanwhile, the Snowflake summit announced notifications in five key areas, including its platform, analytics, data engineering, artificial engineering and machine learning and application and cooperation. Here is a short overview of new things in each category:
Snowflake platform
On the platform side, Snowflake announces its new adaptive computing functionality, which allows Snowflake to manage the sources used for customer questions. Instead of choosing t -shirts for their data songs, customers just choose one size and Snowflake automatically modifies computing behind the scenes to adapt to changing workload.
The new service will save money to customers, says Snowflake EVP Christian Kleinerman.
Snowflake launches a new option for adaptive calculation for your data warehouse (picture with kind permission of Snowflake)
“We improve the price performance of the Snowflake platform and, based on simplification of the allocation of resource calculation, we also help customers better use the computational sources that allocated,” he says. “Start with the adaptive concept of the warehouse and over time we will include the code clusters as a way to simplify and unify our computer models.
Snowflake has also launched the Gen2 data warehouse, which brings twice to 4x power performance performance for-heavy and updating workloads. “We are incredibly excited that with these two combinations we continue to place Snowflake in the foreground not only of performance, but also price performance and the most important, with very easy use,” Kleinerman said.
Data engineering
On the queue of data engineering, Snowflake publishes a new data ingestion offering dubbed openflow. Based on Open Source Apache NIFI and Snowpipe streaming, openflow will allow users to connect to a wide range of structured or non -compatible data such as relational and NOSQL databases, streaming data services such as Kafka or even PDF documents, and bring data to the snowflake.
OpenFlow has a low-code visual interface where customers can create data connection using one of the 30-plus pre-created connectors or build their own connectors. The software will contain a direct acyclic graph visualization (DAG) to help users understand how data flows in their environment. “There is great emphasis on open flow in the topic of visibility, so customers can easily see what they see, what data flows, where data or data quality can exist,” Kleinerman said.
OpenFlow can run as a fully managed service that honors and runs with the guarantee of management and safety, which is built into the snowflake, Kleinerman said. In addition, Snowflake will also support the OpenFow version with its own calculation that can run in the virtual private cloud of customers (VPC) around where customers manage data movement.
“Most of our customers are interested in landing data in a snowflake or making data available for a snowflake, but there are no restrictions as a general frame,” he said. “Our goal is to be able to simplify the movement of data and process data from any source to any other goal.”
Snowflake OpenFlow is a tool for integrating data based on Apache NIFE (picture with kind permission of snowflake)
Better integration with DBT is also on the docket. Soon there will be a public view of the new DBBBBED DBT project that will allow customers to use DBT in Snowflake workspaces, taking advantage of DBT goodies such as code assistance, native Git Integration and side by side. This will follow with the support of DBT Fusion, the new version of the RBT-Power DBT, which was an Andded last week.
Analytics
On the front of AI and machine learning, it uses the Snowflake Summit 25 to start something called CALLEX AISQL, which is a series of extensions for Snowflake’s dialect SQL designed “to perform AI operations as simple as a functional call,” Kleinerman said.
With AISQL, fewer technical users will be able to conjure up AI-Powed features such as the extraction of information from a PDF document or audio file, only a simple API or a functional call, he said.
“What it does brings AI to a wide user base that Snowflake had over the years with a simple and intuitive programming model,” Kleinerman said. “Again, (is) it has tilted very strongly on your data processing platforms, it must be easy and simple.”
Snowflake says that Cortex Aisql will allow customers to create complex multi -stage AI workflows using SQL force. For example, the question as “What is the annual returned growth and market views for companies that have recently undergone a change in the CEO and operated in the renewable energy sector?” It will require a number of steps, including search documents, criteria filtering, key information extraction, connection of non -comprehensive knowledge and synthesizing findings.
“Ai-powred operations, from document filtering to semantic extraction to intelligent connections, all in the well-known SQL syntax, eliminate the need for more specialized tools and its own code,” says Snowflake in the blog. “This unified approach transforms what traditionally requires expertise in the field of data sciences and weeks of development to direct SQL questions that commercial analysts can build and modify in minutes.
Ai and ml
Snowflake Cortex Aisql allows users to call routines AI built into SQL (image with Snowflake permission)
On the front of AI and machine learning, Snowflake has made several notifications, included Snowflake Intelligence, a new application that provides conversational interfaces for business users and data experts to ask natural language questions and answer the ai-nap.
Snowflake Intelligence, which will soon be in a public view, will provide a method wrapped in shrinkage, how to bring Openai and anthropic foundation models to apply against customers. Instead of forcing customers to create their own conversational interface, Snowflake intelligence processes all connections as well as security and data management.
Snowflake also launches preliminary agents. Agent for data science will help automate tedious tasks ml and help solve their workflows ml. Data Science Agent is based on Claude Anthropic and works with ML pipelines developed in the Snowflake laptop.
The company also reveals SnowConvert AI, a new solution for accelerating the migration of competing customer data warehouses on the snowflake. The offer uses Cortex AI to convert code, BI messages, ETL and other components of data warehouse to Snowflake Code, thus relieving data engineers’ burden.
Application and collage
On the queue of Snowflake applications and collaboration, it also produces data sources from data suppliers such as Insights CB, Packt, Stack Overflow, Associated Press and USA Today, which are now available through extension of Cortex knowledge. These will generally be available soon.
Finlly, Snowflake also makes some news in the semantic arena. Semantic models are commonly evolved as part of data modeling efforts within business intelligence or analytics and help ensure the metrics are folded and each on the same page. Because users allow you to share semantic models, it hopes that Snowflake will control some semantic standardization of data shared by the Snowflake Marketplace.
Summit Snowflake Summit takes place on Thursday.
Related items:
Crispy data is all-in with postgres
It’s Snowflake vs. Databricks in a duel of large data conferences
Snowflake Unleashere Agans Agans and Unlock Business Data
(Tagstotranslate) bark