Classmethod Data Analytics Communication (AWS Data Analysis Edition) - August 2025 Issue
I am Ishikawa from the Cloud Business Division Consulting Department. In the summer of 2025, numerous new features and improvements were added to data analysis services at AWS Summit 2025 New York. SageMaker, QuickSight, Redshift, Glue, S3 Vectors and others have received updates that enhance productivity and accelerate AI utilization. Here are some other updates to introduce!
Amazon SageMaker Catalog
New Features & Updates
2025/07/01 - Amazon SageMaker Catalog adds AI recommendations for descriptions of custom assets
A feature has been added to Amazon SageMaker Catalog where Amazon DataZone AI automatically suggests descriptions for custom assets. For structured data such as Iceberg tables in Amazon S3, it uses large language models (LLMs) to generate table summaries, use cases, and column-level descriptions. This reduces manual documentation work, improves metadata consistency, and enables faster discovery of data assets within organizations.
Amazon SageMaker Lakehouse
New Features & Updates
2025/07/16 - Amazon SageMaker streamlines S3 Tables workflow experience
Amazon SageMaker has streamlined the S3 Tables workflow, and integration with Amazon SageMaker Lakehouse is now generally available. You can now create tables, load data, and run queries using the query editor or Jupyter Notebook within SageMaker Unified Studio. You can centrally access data in S3 Tables, S3 buckets, and Redshift warehouses through SageMaker Lakehouse.
Amazon DataZone
API Changes
2025/07/01 - Amazon DataZone - 1 updated methods
Added support for a new optional domain-unit-id parameter to the UpdateProject API.
2025/07/15 - Amazon DataZone - 4 updated methods
Removed restrictions on environment profile ID as a required field and released S3 features.
2025/07/24 - Amazon DataZone - 2 updated methods
- Released highlighting of relevant text in results returned by Search and SearchListings APIs.
- Added support for returning aggregated counts of values for specified attributes in the SearchListings API.
Amazon Redshift / Amazon Redshift Serverless### New Features & Updates
2025/07/16 - Amazon Redshift announces support for cascading refresh of nested materialized views
Amazon Redshift now supports cascading refresh of nested materialized views (MVs). This enables efficient updating of MVs not only on local tables but also on external streaming sources such as Kinesis and MSK.
The new REFRESH command with cascade option updates not only the specified MV but also all dependent lower-level MVs in a single transaction.
For example, when refreshing an MV called v with cascade, its base MV u is refreshed first, followed by v. Conversely, the restrict option limits the refresh to a single MV. This feature makes it easy to maintain data freshness across multiple MVs with dependencies.
CREATE TABLE t(a INT);
CREATE MATERIALIZED VIEW u AS SELECT * FROM t;
CREATE MATERIALIZED VIEW v AS SELECT * FROM u;
CREATE MATERIALIZED VIEW w AS SELECT * FROM v;
-- w -> v -> u -> t
INSERT INTO t VALUES (1);
Amazon Redshift Serverless
New Features & Updates
2025/07/23 - Amazon Redshift Serverless Now Supports 2-AZ Subnet Configurations
Amazon Redshift Serverless now supports subnet configurations with 2 Availability Zones (AZs) when not using Enhanced VPC Routing (EVR). Previously, workgroup creation required 3 AZs, but this requirement has been relaxed.
2025/07/23 - Amazon RDS for Oracle zero-ETL integration with Amazon Redshift
With the zero-ETL integration between Amazon RDS for Oracle and Amazon Redshift, you can leverage petabyte-scale transaction data for analytics and machine learning in near real-time without building complex ETL pipelines. Data written to RDS for Oracle is replicated to Redshift within seconds, enabling quick and comprehensive insights from data across multiple applications.
This feels like "Oracle and Amazon Redshift zero-ETL integration has finally arrived!"
AWS Glue
New Features & Updates#### 2025/07/16 - AWS Glue now supports zero-ETL integrations from Amazon DynamoDB and eight applications to S3 Tables
AWS Glue has started supporting zero-ETL integrations from Amazon DynamoDB and eight applications to Amazon S3 Tables. This automates data extraction and loading from applications such as Salesforce and SAP to S3 Tables.
2025/07/17 - AWS Glue now supports new workers for larger and memory intensive workloads
New worker types have been added to AWS Glue to improve data integration and processing performance. For large-scale, resource-intensive workloads, general-purpose "G.12X" and "G.16X" with enhanced computing, memory, and storage have been introduced. Additionally, memory-optimized workers "R.1X", "R.2X", "R.4X", and "R.8X", which have twice the memory of G workers and are optimal for memory-intensive Spark processing such as caching and aggregation, are now available.
2025/07/24 - AWS Glue now supports Microsoft Dynamics 365 as a data source
AWS Glue has begun providing a new native connector for Microsoft Dynamics 365. This allows data engineers to easily integrate data from Dynamics 365's ERP and CRM platforms as data sources for efficient ETL jobs on AWS Glue.
API Changes
2025/07/23 - AWS Glue - 7 updated methods
AWS Glue has started supporting dynamic session policies for job execution. This allows you to specify fine-grained custom permissions for each job execution without creating multiple IAM roles.#### 2025/07/31 - AWS Glue - 5 updated methods
Root nodes, S3 Iceberg sources/targets, catalog Iceberg sources, DynamoDB ETL connector, AutoDataQuality assessments, enhanced PII detection with redaction, Kinesis fan-out support, and support for new R-series worker types have been added.
Amazon QuickSight
New Features & Updates
2025/07/01 - Amazon QuickSight launches Trusted Identity Propagation (TIP) for Athena Direct Query
Amazon QuickSight has started supporting Trusted Identity Propagation (TIP) for Amazon Athena Direct Query. This feature allows fine-grained control of access permissions to data on QuickSight at the row and column level for each user using AWS Lake Formation rules. As a result, it becomes possible to securely display only authorized data to each user even when the same dashboard is used across multiple customers or departments.
2025/07/2 - Amazon QuickSight supports 2B row SPICE dataset
In Amazon QuickSight Enterprise Edition, the capacity of SPICE datasets has been expanded from the previous 1 billion rows to 2 billion rows, doubling the capacity.
2025/07/09 - Amazon Q in QuickSight is now available in 7 additional regions
Amazon Q in QuickSight is now generally available in new regions in the Asia Pacific area, including Tokyo, Seoul, and Singapore.
This generative BI feature allows users to extract insights from data using natural language, enabling answering questions and automatically generating executive summaries. This accelerates the work of business analysts and supports rapid decision-making.
Additionally, the "Scenarios" feature for Pro users has also expanded in availability. This regional expansion contributes to reducing service latency and addressing data residency requirements, while ensuring high security as customer data is not used for model training.#### 2025/07/09 - Amazon QuickSight introduces granular access customization for exports and reports
Amazon QuickSight has introduced new access control features for data exports and reports. This allows administrators to configure user access permissions in detail for each export type, such as PDF reports, CSV, and Excel files.
This feature applies to both direct exports from dashboards and scheduled email reports. It's also now possible to restrict email reports to only send links to content without attaching files.
2025/07/18 - Unifying data insights with Amazon QuickSight and Amazon SageMaker
At AWS Summit 2025 New York, the integration of Amazon SageMaker and Amazon QuickSight was announced. This collaboration enables seamless transition from data exploration in SageMaker to dashboard creation in QuickSight with just one click. Created dashboards are cataloged as assets within SageMaker projects, making discovery and sharing across the organization easier.
Create a SageMaker domain and enable the QuickSight blueprint.
- Add the QuickSight blueprint to your project profile.
- Log in to SageMaker Unified Studio and create a project with the QuickSight blueprint enabled.
- Select a table from the lakehouse catalog in the project's data tab and execute the "Open in QuickSight" action.
- This single-signs the user into QuickSight and creates a dataset.
- In QuickSight, a "restricted folder" associated with the data source is created, and all assets such as datasets, analyses, and dashboards are stored in that folder.
- Dashboards created in QuickSight automatically become discoverable in the SageMaker Unified Studio project catalog, allowing for sharing and publishing.
API Changes
2025/07/15 - Amazon QuickSight - 3 updated methods
Introduced custom instructions for topics.
2025/07/31 - Amazon QuickSight - 6 updated methods
Added support for Impala connector.
Amazon OpenSearch Service
New Features & Updates#### 2025/07/28 - Amazon CloudWatch and Amazon OpenSearch Service launch pre-built dashboard for AWS Network Firewall
Amazon CloudWatch and OpenSearch Service have announced a new pre-built dashboard for AWS Network Firewall. It provides insights into network metrics such as traffic patterns and protocols, enhancing security teams' troubleshooting capabilities. Administrators can quickly identify abnormal network activity and monitor private network connections to optimize security configurations. This dashboard can be created in CloudWatch Logs Insights or the OpenSearch Service console.
API Changes
2025/07/15 - Amazon OpenSearch Service - 7 updated methods
Support for enabling the S3 Vector engine option has been added to AWS OpenSearch. When this option is enabled, you can create indexes using the S3 Vector engine.
2025/07/31 - Amazon OpenSearch Service - 7 updated methods
Support for fine-grained access control with NEO-SAML using IAMFederation for AOS data sources.
Amazon OpenSearch Serverless
API Changes
2025/07/29 - OpenSearch Service Serverless - 4 updated methods
Support for fine-grained access control with SAML using IAMFederation for AOSS.
Amazon EMR
API Changes
2025/07/31 - Amazon EMR - 3 updated methods
A new parameter "ExtendedSupport" has been added to the Amazon EMR RunJobFlow, ModifyCluster, and DescribeCluster APIs.
Amazon MSK
New Features & Updates
2025/07/16 - Announcing Model Context Protocol (MCP) Server for Amazon MSK
Amazon MSK has announced the Model Context Protocol (MCP) Server, allowing MSK clusters to be operated using standardized natural language interfaces and AI agents. This MCP server provides AI agents with aggregated information about cluster metrics, configuration, and operational status. Agents can understand constraints such as quotas, capacity limits, and best practices to make informed decisions about cluster changes.#### 2025/07/21 - Amazon MSK now supports up to 5x more partitions on Express Brokers
Amazon Managed Streaming for Apache Kafka (Amazon MSK) Express Brokers now support up to 5 times more partitions per broker. This improves price-performance by up to 50% for partition-constrained workloads, allowing operation with fewer brokers. For example, m7g.16xlarge instances now support up to 20,000 partitions, up from the previous 4,000.
Additionally, new upper limits on partition counts have been established to maintain cluster stability. This enhancement is particularly effective for applications handling numerous topics and is available at no additional cost.
AWS Clean Rooms
New Features & Updates
2025/07/01 - AWS Clean Rooms supports incremental and distributed training for custom modeling
AWS Clean Rooms has enhanced its machine learning capabilities with new support for "incremental training" and "distributed training." Incremental training allows building new models based on existing trained models, significantly reducing training time and computational resources. Meanwhile, distributed training enables efficient handling of large datasets by distributing processing across multiple instances.
API Changes
2025/07/01 - AWS Clean Rooms ML - 1 new 16 updated methods
Support for incremental training and distributed training of custom models has been introduced to AWS Clean Rooms ML.
2025/07/17 - AWS Clean Rooms ML - 2 updated methods
Support for Parquet result format for ML input channel models has been introduced to AWS Clean Rooms ML.
2025/08/01 - AWS Clean Rooms Service - 1 updated methods
You can now update table references and allowed columns on existing configured tables.
Amazon S3 Tables
New Features & Updates#### 2025/07/15 - Amazon S3 Tables now supports Model Context Protocol (MCP) Server
AWS announced "Model Context Protocol (MCP) Server for Amazon S3 Tables" to enhance AI-powered data management. This MCP server enables AI code assistants to accurately understand the functions and operations of S3 Tables in context.
Developers and data engineers can use natural language to instruct table creation, schema definition, data import/export, query execution, and can direct and operate data projects using natural language.
API Changes
2025/07/15 - Amazon S3 Tables - 2 updated methods
Adds table bucket type to ListTableBucket and GetTableBucket API operations.
Amazon S3 Vectors
New Features & Updates
2025/07/15 - Announcing Amazon S3 Vectors (Preview)—First cloud object storage with native support for storing and querying vectors
At AWS Summit 2025 New York, a preview of S3's new service "Amazon S3 Vectors" was announced. Amazon S3 Vectors is a cost-optimized vector storage for AI agents and semantic search. It reduces the cost of storing and querying vector data by up to 90%, making it easier to utilize large datasets to improve AI context understanding and search results.
While maintaining the same scalability and durability as S3, it delivers high-speed queries in sub-second time. It integrates with Amazon Bedrock's RAG (Retrieval-Augmented Generation) and OpenSearch Service, enabling flexible data management with lower costs. Infrastructure management is not required thanks to dedicated APIs, and it can be used for various purposes such as video search and pattern detection in medical images.
API Changes
2025/07/16 - Amazon S3 Vectors - 16 new methods
Amazon S3 Vectors provides cost-effective, elastic, and durable vector storage for queries based on meaning and similarity.## Finally
At AWS Summit 2025 New York, there were numerous new services and major updates, including the preview of a new S3 service called "Amazon S3 Vectors" and the integration of SageMaker with Amazon QuickSight. Additionally, AWS was actively providing various MCP servers amid the expansion of AI code assistants such as Amazon Q Developer.
We recommend that you review the new services and updates announced at AWS Summit in New York 2025.
![[小ネタ]S3 レプリケーションが失敗する原因を突き止めるために S3 イベント通知を設定してみた](https://images.ctfassets.net/ct0aopd36mqt/wp-thumbnail-066beb776f0c57ce64255fadcc072f60/82b2f6687ab5774cd73f9176dcac7855/amazon-s3)
![[アップデート] CloudFrontの署名付きURLでECDSA鍵を利用できるようになりました！](https://images.ctfassets.net/ct0aopd36mqt/wp-thumbnail-f66629e23bafcf47fa6477fd0349a4f7/3e371b05917375d87daf613d54a600d8/amazon-cloudfront)
