The Unofficial Snowflake Monthly Release Notes: November 2024

Monthly Snowflake Unofficial Release Notes #New features #Previews #Clients #Behavior Changes

Welcome to the fantastic Unofficial Release Notes for Snowflake for November 2024! You’ll find all the latest features, drivers, and more in one convenient place.

As an unofficial source, I am excited to share my insights and thoughts. Let’s dive in! You can also find all of Snowflake’s releases here.

This month, we provide coverage up to release 8.44 (General Availability — GA). I hope to extend this eventually to private preview notices as well.

I would appreciate your suggestions on continuing to combine these monthly release notes. Feel free to comment below or chat with me on LinkedIn.

Behavior change bundles 2024_06 are active by default, 2024_07 is enabled by default but can be disabled, and 2024_08 can be enabled. 2024_08 is a security heavy bundle with things like enforcing new password sizes.

What’s New in Snowflake

New Features

• Snowflake Connector for SharePoint (Preview) connects a Microsoft 365 SharePoint site and Snowflake to ingest files and user permissions and keeps them up to date. Snowflake Connector for SharePoint also supports the Cortex Search service and can make ingested files ready for conversational analysis for use in AI Assistants using SQL, Python or REST APIs.
• Outbound private connectivity for Snowflake features, create private endpoints in Snowflake to access a cloud platform using the platform’s private connectivity solution rather than the Internet.
• External stages using Azure Private Link (Preview), configure an external stage and create a private endpoint so bulk loading from Azure storage occurs over Azure Private Link.
• External volumes using Azure Private Link (Preview), configure an external volume and create a private endpoint so you can connect Snowflake to your external cloud storage for Iceberg tables using Azure Private Link instead of the public Internet.
• Snowpipe automation using Azure Private Link (Preview), configure an external stage and notification integration, and create a private endpoint, so that automatic Snowpipe data loads that are triggered by Microsoft Azure Event Grid use Azure Private Link instead of the public Internet.
• Visual Studio Code extension for Snowpark Python (GA), the extension now integrates with Snowpark Python to provide authoring and debugging features for Snowpark Python code. These new features include Inline debugging of Snowpark Python functions, Syntax highlighting and autocomplete suggestions for Snowflake SQL in Python strings within Python files or notebook cells, and Syntax highlighting and bracket autocomplete of Jinja templates in Snowflake SQL.
• Tasks: Python and JVM support for serverless tasks(GA), invoke the following object types and functions: UDFs (user-defined functions) and stored procedures written in Python, Java, and Scala.
• Leaked password protection is a background service in Snowflake that monitors and disables leaked passwords to help prevent unauthorized access to Snowflake accounts. The service also provides a notification system for administrators so they are aware of leaked passwords when they are detected in external databases.
• Serverless alerts (GA), a serverless compute model for Snowflake alerts, Snowflake determines the ideal size of the compute resources for a given run based on a dynamic analysis of statistics for the most recent previous runs of the same alert.
• Replication error notifications for replication and failover groups (GA): Set a notification integration for a primary replication or failover group to receive error notifications for refresh operation failures.
• Hybrid tables support extended to additional AWS regions: Europe (Zurich) eu-central-2 and Asia Pacific (Jakarta) ap-southeast-3
• Manage account preview features (GA): Account administrators can enable, disable, and view the status of preview features within their accounts.
• Logical replication of clones(GA), when the original table and cloned table are included in the same replication or failover group, the cloned table can be replicated logically to the target account. As a result, logical replication, versus physical replication, reduces egress and replica storage costs.

Snowsight Updates

• Snowflake Notebooks Warehouse Runtime (GA) on Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) commercial regions. Snowflake Notebooks is a development interface in Snowsight that offers an interactive, cell-based programming environment for Python and SQL. In Snowflake Notebooks, you can perform exploratory data analysis, develop machine learning models, and perform other data science and data engineering tasks all in one place.
• Snowsight rate limits (GA) enhance platform security and performance under high traffic conditions, such as DDoS attacks or unexpected surges in request volume. Controlled limits are posed on the number of requests that can be made to Snowsight within specified time frames, ensuring consistent availability and stability across all user sessions.

Snowflake Applications

• Snowflake Native Apps with Snowpark Container Services in AWS (GA) can be distributed to any Snowflake customer who can use them in production AWS commercial regions. Apps with containers provide all of the functionality of Snowpark Container Services, including compute pools, services, jobs, external access integrations, etc. within a Snowflake Native App.
• Snowflake Native Apps with Snowpark Container Services in Azure (Preview) can be distributed to any Snowflake customer who can use them in Azure commercial regions.
• Native App Framework support for Budgets a Snowflake Native App can use Budgets to monitor credit usage. Customers can set up spending limits for app, which include all the app’s compute resources, including compute pools and warehouses. After installing an app, consumers can view and create budgets.
• Snowflake Native Apps: Multiple app installs (GA), allows consumers to install multiple instances of an app in their account.

Data Lake Updates

• Apache Iceberg™ tables: Support for Microsoft Fabric OneLake storage (Preview), create an external volume that connects Snowflake to Fabric OneLake storage, then create a Snowflake-managed table that writes to that location. You can query the table using both Snowflake and Fabric.
• Specify an external ID for SIGV4 REST catalog integrations, support for specifying an external ID when you create a catalog integration for Apache Iceberg™ REST that uses SIGV4 authentication. Specifying an external ID lets you use the same IAM role across multiple catalog integrations, which is helpful in testing scenarios when you need to recreate or replace a catalog integration many times.
• Dynamic tables: Support for reading from Snowflake-managed Iceberg tables and creating dynamic Apache Iceberg™ tables (GA). You can create a dynamic table that reads from a Snowflake-managed Iceberg table as the source, just like regular tables. You can create a dynamic Iceberg table, a new dynamic table type that stores query results in the Iceberg table format.
• Apache Iceberg™ tables: Efficient bulk loading, continuous ingestion, and data streaming (GA) use the same core Snowflake ingestion features like COPY INTO
  , Snowpipe, and Snowpipe Streaming, to load data into both standard Snowflake tables and Iceberg tables.

  S3-compatible storage for externally managed Apache Iceberg™ tables(GA): use an external volume to connect to and query Iceberg tables in an S3-compatible storage location.

  SQL Updates

  • PARSE_JSON and TRY_PARSE_JSON functions, duplicate keys are now permitted as a parameter argument, and when set to “d,” only the value of the last occurrence of each key is returned.
  • EXECUTE IMMEDIATE FROM: Support for using content from staged files in templates, in a Jinja2 template, you can include, import, inherit from, and read content from other files on a stage, which enables you to make your templates more modular.
  • Automatic logging and tracing for Snowflake Scripting stored procedures. The additional log information includes the BEGIN/END of a Snowflake Scripting block and a child job request. The additional types of trace events include exception catching, information about child job execution, child job statistics, and stored procedure statistics, including execution time and input values. You can generate this additional information without modifying the body of the stored procedure.
  • CCOUNT_USAGE: New SERVERLESS_ALERT_HISTORY view, query this view to get information about the credits used for serverless alerts.
  • Grouped Query History (Preview), a view in Snowsight to monitor usage and performance of critical and frequently run queries. This graphical view is based on information that is recorded in the AGGREGATE_QUERY_HISTORY view and is particularly useful for monitoring and analyzing Unistore workloads.
  • Additional CREATE OR ALTER commands (Preview) with these commands combine the functionality of the CREATE command and the ALTER command. It provides a declarative and idempotent approach to defining your Snowflake objects. When used together with the Git integration, this enables an Infrastructure-as-Code (IaC) approach to database change management. The following additional objects are supported: APPLICATION ROLE, DATABASE, DATABASE ROLE, ROLE, SCHEMA, STAGE, VIEW, WAREHOUSE.

  AI & ML updates

  • API-level Role-based Access Control (RBAC) for Cortex Analyst requires all requests made to Cortex Analyst to use a role that has been granted the CORTEX USER role. This provides admins with a way to control who can call Cortex Analyst with Snowflake RBAC. CORTEX_USER is granted to PUBLIC by default.
  • Full-text search (GA), use full-text search, call the new SEARCH and SEARCH_IP functions to find character data (text) and IP addresses in specified columns from a table, including elements in VARIANT, OBJECT, and ARRAY columns.
  • Top Insights ML Function for key driver analysis (GA), easily identify drivers of a metric’s change over time or explain differences in a metric among various verticals. It is integratable into business intelligence workflows powering business reviews, business dashboards, and anomaly detection tools to understand key drivers impacting various metrics.
  • Cortex AI CLASSIFY_TEXT function (GA) is purpose-built to help you easily classify text records such as emails, call transcripts, and product reviews into categories relevant to your business.
  • SPLIT_TEXT_RECURSIVE_CHARACTER Cortex function (Preview), the function splits a string into smaller chunks of text, recursively, so that the text can be passed to embedding or search indexing functions. Since many language models have a limit on the number of tokens they can process, this function is essential to processing text larger than the token limit.
  • Snowflake ML Classification(GA) is a machine learning function that sorts data into different classes using patterns detected in training data, making it easy for data scientists and analysts to quickly get binary or multi-class predictions. Top use cases for Snowflake ML Classification include powering buy and churn predictions, credit card detection, and spam detection.
  • Snowflake ML: Distributed Hyperparameter Optimization on Snowpark Container Services (Preview) The ML Hyperparameter Optimization (HPO) API is a model-agnostic framework that enables efficient, parallelized hyperparameter tuning of models. This API is available within a Snowflake Notebook configured to use the Container Runtime on Snowpark Container Services (SPCS).
  • Cortex Analyst feature — Multi-turn conversation in Cortex Analyst (Preview), enables asking follow-up questions that build on previous queries, creating a more dynamic and interactive data exploration experience. For example, the user asks: “What is the month-over-month revenue growth for 2021 in Asia?”, and then follows up with: “What about North America?”
  • Cortex Analyst — Joins support in Cortex Analyst (Preview), supports SQL joins, enabling more advanced data analysis across multiple tables, especially in star schema structures. This feature allows you to query data from fact tables and associated dimension tables with ease.
  • Updates to the Snowflake Cortex TRANSLATE function: The TRANSLATE function provides high-quality, reliable translations for call transcripts, product reviews, social media comments, and other text. This includes improved translation quality, Improved translation reliability, Longer context length (up to 4,096 tokens), new languages (Dutch, Chinese, and Hindi), and Mixed language support. Text written in a mixture of two languages can now be translated into a single language.
  • Snowflake ML: Model Observability (Preview), includes monitoring model behavior over time.

  Data Clean Rooms Updates

  • All developer API clean rooms are now available in the web app, allowing users to manage these clean rooms directly in the web app while still being able to run any template with a custom web app form. Providers no longer need to call the register clean room to web app APIs, in order to make clean rooms available in the web app.
  • Provider run for custom web app templates, enable provider run on custom web app templates. This enables consumers to install and set their respective policies directly through the web app, while allowing the provider to configure and execute the template query via the web app as well. Providers must request provider run to be enabled via developer APIs and then call the create or update listings API, prior to the consumer installing this in the web app. Additionally, providers can customize web app form drop-downs to reference options for consumer join & column policies.
  • Provider and consumer activation in custom web app templates, add a custom activation template to their custom analysis template in the web app. This enables collaborators to support activation use cases, while deploying custom analysis templates within the web app. Providers will need to add a reference to their activation template in the web app form.
  • SQL policy configuration updates and join columns will be selected by default for both aggregation and projection policies. Users can remove and customize their policy requirements as they see fit, while no longer being required to add a join column for every table in the clean room.
  • Sync and naming support for data connectors, manually sync their data connectors to reflect any changes in the metadata related to the table in the web app. Additionally, users can provide their preferred name for these external tables, which is prefixed with the cloud identifier for ease of reference.
  • Non-overlap metrics: Clean room statistics now include non-overlap metrics when using the Audience Overlap & Segmentation template in the web app. This shows how many records in your data did not match join IDs in the collaborator’s data. This capability must be enabled by the data provider.
  • Unlink datasets API: unlink datasets that were previously linked to the clean room using the API.
  • Dynamic table support, users can now register and use dynamic tables in their clean rooms. To register these objects via the APIs, see library.register_objects in the consumer or provider documentation.
  • Custom Python code in consumer templates: Create a custom template that can now upload and reference custom Python code in their template.
  • Merkury Identity connector is an identity service provided by Merkle to Snowflake Data Clean Room customers so that they can encode their customer identifiers into encoded Merkury IDs prior to collaborating within the clean room. The Merkury Identity connector supports multiple type of IDs: HMID (hashed Merkury ID), email (in clear-form, or encoded in MD5, SHA1 or SHA256), device ID, IP address, and phone (only supported in cleartext).
  • Google Display & Video 360 — Customer Match activation connector, which pushes your first-party, custom-audience data into your Google DV360 account.

  Data Pipelines/Data Loading/Unloading Updates

  • Tasks: Serverless tasks user control(GA), take some control over the cost and performance of serverless tasks by setting the following parameters: SERVERLESS_TASK_MAX_STATEMENT_SIZE, SERVERLESS_TASK_MIN_STATEMENT_SIZE, and TARGET_COMPLETION_INTERVAL.
  • Tasks: Task success notifications (GA) push success notifications to a cloud messaging service when a task graph, not individual DAGs, completes successfully.
  • Dynamic tables: Support for replication across different failover groups

  Performance Updates

  • Top-k pruning for queries that contain aggregate functions, expands top-k pruning to include queries that contain aggregate functions.

  Security & Governance Updates

  • Trust Center: Two new scanners in the Security Essentials scanner package (Multi-Factor Authentication (MFA) Enforcement and Event Table Configuration for Native Apps logging verification).
  • Data Lineage (Preview) is a feature that automatically tracks the flow of data between Snowflake objects in real-time, for example, from a table to a view. Relationships among table-like objects, columns, and stages are supported, as well as between data objects and machine learning objects including datasets, feature views, and models. You can use lineage information to assist in impact analysis, monitoring, troubleshooting, and compliance efforts. Lineage can also help you propagate knowledge of sensitive data elements using tags, and it is available through Snowsight, SQL, and Python.
  • Budgets: Support for cloud provider queue and webhook notifications to a queue provided by a cloud service (Amazon SNS, Azure Event Grid, or Google Cloud PubSub), a webhook for Slack, Microsoft Teams, or PagerDuty.
  • Automatic Sensitive Data Classification (Preview) is a serverless feature that can help detect sensitive data using native and custom classifiers. It can also automatically apply user-defined tags and masking policies to columns when sensitive data is detected.
  • Governance for organization listings through access history, organization-level access history has been enhanced with columns that provide information about how data provided by organizational listings is being queried by consumers.

  Extensibility updates

  • The new TensorFlow version 2.17.0 introduces a revised module structure for Keras, a deep-learning API. If your UDFs or procedures don’t specify an earlier TensorFlow version, Snowflake will default to using 2.17.0 when you execute CREATE OR REPLACE.
  • Authentication with AWS IAM from procedures and functions (GA): Support for authenticating with AWS services from a procedure or function using Snowpark External Access via Identity and Access Management (IAM).
  • External network access for Azure Gov regions(GA), access network locations external to Snowflake from within procedure, and UDF handler code is now generally available in all regions.

  Listings updates

  • LISTING_REFRESH_HISTORY (GA), use this function to view the past 14 days of refresh history for a cross-cloud auto-fulfillment listing. The information returned contains replication details for refresh events where the listing is synchronized to a specified target region.
  • Organizational listings and the Internal Marketplace(GA) is a directory of data products shared across the customer’s organization. It lets customers discover and access available data and app products from all teams and business units within their Snowflake Organization.

  Open-Source Updates

  • terraform-snowflake-provider 0.98.0 (New Features: Resources: Added authentication_policy, external_volume, stream_on_*, primary/secondary_connection, and secret types, Data Sources: Introduced connections and secrets, Reworks: Improved provider configuration and streams data source, SDK: Upgraded tag SDK and enabled stale stream recreation, Miscellaneous: Object renaming research/tests. Updated roadmap and minor fixes, Bug Fixes: fixed grant/user imports, stream behavior, and resource creation logic, resolved key issues and improved test coverage)
  • terraform-snowflake-provider 0.99.0 (New Features: tags data source (#1372) and Tag resource V1 (#1806, #1443, #1394), Tasks V1 readiness, Misc: version usage tracking (#3224), improved tags and storage integration tests (#3193, #3213), unskipped and enhanced auth, parser, and secret tests, Bug Fixes: minor fixes (#3226))
  • Snowflake VS Code Extension 1.10.5 (fixed SQL compilation errors on login, upgraded to NodeJS Driver 1.15.0)
  • Streamlit 1.40.1 (small bug fixes and improvements)
  • Streamlit 1.40.2 (small bug fixes and improvements)

  Client, Drivers, Libraries and Connectors Updates

  New features:

  • Snowflake Connector for SharePoint 1.0.0 (initial release)
  • .NET Driver 4.2.0 (added a signature on the driver package to verify its authenticity and integrity, support for reading vector types. For more information, see VectorType.md, support for reading structured types for the JSON result format, logging for the client environment configuration, implemented SnowflakeDbDataReader.GetEnumerator().)
  • Ingest Java SDK 3.0.0 (Snowpipe Streaming can ingest data into Snowflake-managed Apache Iceberg tables)
  • Node.js 1.15.0 (support for Node.js version 22, checks for the PROXY* (such as proxyHost) and the noProxy environment variables when creating an httpAgent, support for the describeOnly configuration parameter, improved logging at the connection layer)
  • ODBC 3.5.0 (support for Red Hat Enterprise Linux (RHEL) 8 and CentOS 8 for ARM64 processors, connectivity diagnostics mode, including the following new connection parameters: enable_connection_diag, which controls whether the connector generates a connectivity diagnostic report, connection_diag_log_path, which is the absolute path where the connectivity report is stored, connection_diag_allowlist_path, which is the absolute path to a JSON file containing the output of SYSTEM$ALLOWLIST() or SYSTEM$ALLOWLIST_PRIVATELINK(), updated the following libraries: curl from version 8.7.1 to 8.10.1, openssl from version 3.0.13 to 3.0.15)
  • Snowflake CLI 3.2.0 (support for event sharing in Native App project definitions, new telemetry section to the application entity, added the following fields to the telemetry section: share_mandatory_events and optional_shared_events, new options to several snow commands: snow sql: added the — retain-comments option to support passing comments to Snowflake, snow object create: added the — replace and — if-not-exists options to support overwriting exist objects, snow stage copy: added the — recursive option to support copying local files and subdirectories to a stage, including glob support, snow app version create: added the — label option to support adding labels to versions and patches, snow connection add: added the — no-interactive option to skip interactive prompts for unspecified parameters, snow spcs service logs: added the following options to improve log retrieval and monitoring: — since: Start log retrieval from a specified UTC timestamp, — include-timestamps: Include timestamps in log entries for log streaming, — follow: Stream logs in real-time, — follow-interval: set custom polling intervals during log streaming, — previous-logs: Retrieve logs from the last terminated container, snow helpers v1-to-v2 command now converts v1 template references to v2 references in Native App artifacts that use the templates processor, updated the snow — info command to return information about the SNOWFLAKE_HOME variable)
  • Snowflake Python API 1.0.2 (removed the async_req parameter (asynchronous mode) from the execute_job API in the Service resource)
  • Snowflake Python API 1.0.1 (support for the following new resources: Cortex Chat, Cortex inference, added support for customized user agents)
  • Snowpark Library for Python 1.25.0 (added the following new functions in snowflake.snowpark.dataframe: map, improvements: when target stage is not set in profiler, a default stage from Session.get_session_stage is used instead of raising SnowparkSQLException, allowed lower case or mixed case input when calling Session.stored_procedure_profiler.set_active_profiler, added distributed tracing using open telemetry APIs for action function in DataFrame: cache_result, removed opentelemetry warning from logging, added a dependency on protobuf>=5.28 and tzlocal at runtime, added a dependency on protoc-wheel-0 for the development profile, require snowflake-connector-python>=3.12.0, <4.0.0 (was >=3.10.0).)
  • Snowpark Library for Python pandas API 1.25.0 (support for Index.to_numpy, support for DataFrame.align and Series.align for axis=0, support for snowflake.snowpark.functions.window, support for pd.read_pickle (Uses native pandas for processing), support for pd.read_html (Uses native pandas for processing), support for pd.read_xml (Uses native pandas for processing), support for aggregation functions “size” and len in GroupBy.aggregate, DataFrame.aggregate, and Series.aggregate, support for list values in Series.str.len, Improvements: np.where with scalar x value by eliminating unnecessary join and temp table creation, get_dummies performance by flattening the pivot with join)
  • Snowpark Library for Python local testing 1.25.0 (support for patching functions that are unavailable in the snowflake.snowpark.functions module, support for snowflake.snowpark.functions.any_value)
  • Snowflake Connector for Google Analytics Aggregate Data 2.2.1 (Behavior changes: Event sharing is now mandatory for all event types)
  • Snowflake Connector for Google Analytics Raw Data 1.7.2 ()
  • Snowflake Connector for Google Analytics Raw Data 1.6.3 ()
  • Snowflake Connector for ServiceNow® V2 5.14 (Behavior changes: Event sharing is now mandatory for new installations. New features: set a specified table page size with the RESET_PAGE_SIZE procedure instead of using the default connector’s value. If the connector’s default page size was set to an invalid value, the connector will use the recommended value of 10,000)
  • Snowflake Connector for PostgreSQL/MySQL 6.8.0 (behavior changes: increased the minimum schedule interval to 15 minutes, pre-existing 10-minute intervals remain unchanged, operational warehouse will now suspend when there is no data traffic from the source database for more than 5 minutes, Killing agent on shutdown container error message will not show when triggering the agent’s shutdown. Instead, the following message will be logged on the INFO level: Stopping the agent on the container shutdown signal, PUBLIC.REMOVE_TABLES(DATA_SOURCE_NAME STRING, SCHEMA_NAME STRING, TABLE_NAMES ARRAY) procedure enables the removal of multiple tables with one call)
  • SQLAlchemy 1.7.0 (support for the following features: Dynamic Tables, Hybrid Tables, Iceberg Tables with the Snowflake Catalog, support for the MAP data type, added the ability to define options in key arguments instead of arguments, updated the cluster_by option to support explicit expressions)

  Bug fixes:

  • .NET Driver 4.2.0 (changed the SnowflakeDbConnection finalizer to be non-blocking, fixed an issue where some disposable objects were not properly disposed, improved memory management for reading large query results, increased the log level of messages for failed HTTP responses, stopped retrying non-recoverable authentication exceptions, fixed a concurrency issue with initializing a connection pool, changed DateTime.Kind to Unspecified for reading DATE, TIME and TIMESTAMP_NTZ Snowflake types. Version 2.1.3 of the driver introduced an undesired change of setting the DateTime.Kind to Utc, fixed null response handling for PUT/GET operations in the GCS client, fixed exception handling for PUT/GET operations in the S3 client, fixed very large or very small timestamps handling, improved the logic for calculating when the next retry will happen, fixed the returning rows count for COPY statements from multiple files, fixed support of PUT/GET files without client side encryption)
  • Ingest Java SDK 3.0.0 (Fixed dependency issues and error messages in the SDK)
  • Node.js 1.15.0 (fixed an issue where the driver did not handle the rejected state of the Promise object in the heartbeat method)
  • Snowflake CLI 3.2.0 (removed the requirement for an existing requirements.txt file for Python code executed with the snow git execute command. Previously, the file must have existed, even if empty, for the command to succeed, removed the requirement for needing a privilege to create a table or schema to execute the snow app version create command if the schema and table already exist, fixed an issue relating to configuration file updates when the connection.toml file exists, no longer incorrectly copying connections from connections.toml to confg.toml files, fixed an issue where the snow connection generate-jwt command failed with keys without a passphrase, fixed a Windows permissions error for file created by Snowflake CLI when the owneris part of a custom group with granted default permissions)
  • Snowflake Python API 1.0.1 (fixed the ValueError message for Enum types, fixed the API documentation for Enum types to show possible values, added the missing DeleteMode type to the API documentation)
  • Snowpark Library for Python 1.25.0 (fixed the pre-action and post-action query propagation when In expressions were used in selects, fixed a bug that raised error AttributeError while calling Session.stored_procedure_profiler.get_output when Session.stored_procedure_profiler is disabled)
  • Snowpark Library for Python pandas API 1.25.0 (fixed a bug where aggregating a single-column dataframe with a single callable function (e.g. pd.DataFrame([0]).agg(np.mean)) would fail to transpose the result., fixed bugs where DataFrame.dropna() would: treat an empty subset (e.g. []) as if it specified all columns instead of no columns, raise a TypeError for a scalar subset instead of filtering on just that column, raise a ValueError for a subset of type pandas.Index instead of filtering on the columns in the index, creation of scoped read-only tables to mitigate TableNotFoundError when using dynamic pivot in a notebook environment, fixed a bug when concat dataframe or series objects are coming from the same dataframe when axis = 1)
  • Snowpark Library for Python local testing 1.25.0 (fixed a bug where Table.update could not handle VariantType, MapType, and ArrayType data types, fixed a bug where column aliases were incorrectly resolved in DataFrame.join, causing errors when selecting columns from a joined DataFrame, fixed a bug where Table.update and Table.merge could fail if the target table’s index was not the default RangeIndex)
  • Snowflake Connector for Google Analytics Raw Data 1.7.2 ()
  • Snowflake Connector for ServiceNow® V2 5.14 (Ingestion fails when a worker task reaches API timeout when discovering the initial table page size)
  • Snowflake Connector for PostgreSQL/MySQL 6.8.0 ( operational warehouse will now correctly suspend when all data sources are switched to a scheduled mode, the agent will no longer restart when the logging rate is too high, fixed failing table replication on PostgreSQL numerics NaN, Infinity, -Infinity by ingesting them as null, fixed replication errors when the DEFAULT_DDL_COLLATION parameter for account is set)
  • SQLAlchemy 1.7.0 (fixed the SAWarning when registering functions with existing name in the default namespace)

  Conclusion

  In the last month, Snowflake’s platform underwent some good amount of changes that brought many AI and ML features into general availability. These improvements not only make Snowflake a more flexible and reliable data platform, but they also make it easier and safer for users to handle, analyze, and gain insights from their data.

  They enabled the Microsoft ecosystem by adding a SharePoint connector and Azure Private Link, Snowflake improves the security and accessibility of data, making it easier to import and process. The platform’s focus on AI and machine learning, along with features like full-text search, advanced ML functions, and better model observability, lets businesses use advanced analytics and prediction tools right from the platform.

  Automatically classifying private data, better tracking of data lineage, and better role-based access controls all improve security and governance in a big way. This makes sure that data compliance and security stay at the highest levels. FYI, that bundle 08 that disabled by default has a lot of changes for security as well. The changes that have been made to Snowflake’s extensibility and open-source contributions show that the company wants to create a collaborative environment where community-driven improvements and integrations can work well.

  Enjoy the reading.

  I am Augusto Rosa, VP of Engineering for Infostrux Solutions. I am also a Snowflake Data Super Hero and Snowflake SME. You can follow me on LinkedIn.

  Subscribe to Infostrux Medium Blogs https://medium.com/infostrux-solutions for the most interesting Data Engineering and Snowflake news.

  Sources:

  • https://docs.snowflake.com/en/release-notes/preview-features
  • https://docs.snowflake.com/en/release-notes/new-features
  • https://docs.snowflake.com/en/release-notes/sql-improvements
  • https://docs.snowflake.com/en/release-notes/performance-improvements-2024
  • https://docs.snowflake.com/en/release-notes/clients-drivers/monthly-releases
  • https://docs.snowflake.com/en/release-notes/connectors/
  • https://marketplace.visualstudio.com/items?itemName=snowflake.snowflake-vsc