This is the second article in our re :Invent 2025 series, following our deep dive into the evolution of compute and the game-changing Lambda Managed Instances announcement here. Today, we shift our focus to the data and database layer, an area where AWS unveiled some of the most financially impactful and architecturally strategic updates of the year.
Among the new capabilities, the introduction of AWS Database Savings Plans stands out as a transformational change that will significantly influence how organizations optimize and scale their database workloads going into 2026.
But this wasn’t the only data-centric news. AWS also announced key changes on how pricing works for MySQL and Oracle RDS, as well as optimizations for across S3 tables, Aurora, DynamoDB, caching layers, and integrations aimed at improving scalability, performance, and operational simplicity. Together, these updates position AWS databases to better support AI-driven applications, event-centric architectures, and high-throughput systems.
Database Savings Plans
Databases often represent one of the largest, most consistent cost centers in an AWS environment. With Savings Plans, organizations can dramatically improve predictability and reduce long-term spend.
For customers with predictable, steady-state workloads, Reserved Instances (RIs) have historically been a popular way to optimize cloud costs. By committing to a specific instance type and size for a 1–3 year term, organizations could achieve significant discounts compared to on-demand pricing.
However, this model also introduced a fundamental limitation. RI commitments are tightly coupled to a specific instance family, engine, and configuration. When demand changes, workloads need to scale differently, or teams want to modernize—such as migrating from RDS to Aurora or moving to a non-relational database—those commitments can quickly become restrictive and reduce architectural flexibility.
Database Savings Plans, provide that flexibily extending the commitment-based discount model, l—previously available for compute—to a wide set of managed database and data services. Customers commit to a specific hourly spend for 1 year in exchange for automatic discounts across eligible services.
Key benefits include:
- Discounts of up to ~35% are now achievable depending on service type and usage patter
- Flexibility across instance sizes, regions, and database engines. (Migrate from RDS to Aurora without losing discounts.)
- Discounts applied automatically to any qualifying database usage.
- No lock-in to specific instance families or configurations.
These Savings Plans apply to a broad set of data services, including:
- Amazon RDS (all engines)
- Amazon Aurora (provisioned & serverless)
- Amazon DynamoDB
- Amazon ElastiCache
- Amazon DocumentDB
- Amazon Neptune
- Amazon Keyspaces
- Amazon Timestream
- AWS Database Migration Service (DMS)
This breadth makes the Savings Plans model a powerful financial tool for cloud architects and FinOps teams.
Another great feature of savings plans is that it can be applied at the organizational level, so companies on a multi-account model, can have greater flexibility and swift savings from legacy workloads into new ones that run on a separate account in their account org.
RDS Updates
Building on the same themes of cost optimization and architectural flexibility introduced with Database Savings Plans, AWS also delivered several important updates to Amazon RDS that directly impact how teams modernize, license, and scale their relational database workloads.
Following on the cost optimization trend, there some of the main announcements include:
- Support for SQL Server Developer Edition, allowing customers to get the full EE functionality on their dev and test workloads without incurring the high licensing cost
- M7i/R7i RDS instances; these new instance types not only provide a 15% increase in performace compared to previous generations, but have 2 key features that could cut in half the current price of the MS SQL and Oracle workloads:
- Unbundled pricing: the new generation of instances, allow the cost of the compute resources (priced per CPU hour) to be separate from the Licensing cost (per vCPU hour)
- Optimiza CPU, by allowing customers to chose the number of thread per CPU Cores used by the instance, it provides better licensing cost by still having the right memory and IOPS configuration and flexibitily
- Additional storage: RDS now supports 256TB of storage, this has been a limitation that stopped multiple customers from moving their datawarehouse and heavy data workloads using Oracle or MSSQL to the cloud.
S3 Tables
S3 Tables received important new capabilities that significantly improve cost efficiency, durability, and multi‑region data strategies for analytics and AI workloads.
- Intelligent‑Tiering support for S3 Tables – allowing table data to automatically move between storage tiers based on access patterns, reducing costs without requiring manual lifecycle policies.
- Native replication support – enabling managed, continuous replication of S3 Tables across AWS Regions and accounts without custom pipelines or tooling.
These enhancements reduce the operational complexity typically associated with large‑scale data lakes while improving resiliency and global availability.
Together, Intelligent‑Tiering and replication bring S3 Tables closer to enterprise‑grade data platform expectations:
- Lower storage costs through automatic tiering based on real usage patterns
- Simpler global data architectures for analytics, disaster recovery, and compliance
- Reduced operational overhead compared to custom ETL‑driven replication models
- Stronger alignment with AI and analytics workloads , where large volumes of historical data are accessed unevenly
L
AWS Glue Zero-ETL & DMS Enhancements
AWS re:Invent 2025 delivered important updates that simplify how organizations integrate, migrate, and modernize data, reducing both operational overhead and long-term cost.
AWS Glue Zero-ETL for Self-Managed Databases
AWS expanded AWS Glue Zero-ETL to support self-managed database sources , including Oracle, SQL Server, MySQL, and PostgreSQL running on-premises or on EC2. This capability enables near real-time replication of transactional data into Amazon Redshift without building or maintaining custom ETL pipelines.
Key benefits include:
- Eliminating custom ETL development and maintenance
- Keeping analytics platforms continuously in sync with operational systems
- Accelerating time-to-insight for BI, analytics, and AI/ML workloads
This makes Zero-ETL a practical option not only for cloud-native systems, but also for hybrid and legacy environments.
AWS DMS Schema Conversion: SAP Sybase ASE → PostgreSQL
In parallel, AWS announced new AWS Database Migration Service (DMS) schema conversion capabilities to help organizations modernize away from legacy databases.
AWS DMS now supports automated schema conversion from SAP Sybase ASE to PostgreSQL , significantly reducing the complexity, risk, and manual effort typically involved in these migrations.
This enhancement enables teams to:
- Migrate off legacy, high-cost database platforms
- Adopt open-source engines like PostgreSQL faster
- Reduce manual schema refactoring and testing
- Align database modernization with broader cost-optimization initiatives such as Database Savings Plans
Together, Glue Zero-ETL and DMS schema conversion reinforce AWS’s broader data strategy: making data easier to move, integrate, and modernize—while lowering both operational and financial barriers.
These advancements reflect AWS’s continued investment in making cloud-native, high-performance data architectures more cost-efficient and easier to operate.
Other Notable Mentions
Beyond the major announcements covered above, AWS re:Invent 2025 included several additional updates worth noting for teams building modern, data-driven platforms.
Amazon Aurora Global Database: Fully Managed Blue/Green Deployments
AWS introduced fully managed blue/green deployments for Amazon Aurora Global Database , enabling safer, faster, and more controlled database updates across multi-region architectures.
Amazon DynamoDB: Multi-Attribute Composite Keys for GSIs
AWS announced support for multi-attribute composite keys on Global Secondary Indexes (GSIs) in Amazon DynamoDB. This enhancement allows developers to model more complex access patterns without duplicating data or creating excessive secondary indexes.
Aurora DSQL Cluster Creation in Seconds
AWS introduced the ability to create Amazon Aurora Distributed SQL (Aurora DSQL) clusters in seconds , dramatically accelerating the time it takes to provision globally distributed databases. This enhancement enables faster experimentation, failover testing, and multi-region deployment strategies for mission-critical applications.
🔗 https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-aurora-dsql-cluster-creation-in-seconds/
Aurora PostgreSQL Integration with Kiro Powers
AWS announced Aurora PostgreSQL integration with Kiro Powers , expanding data access and operational capabilities. This integration enhances how Aurora PostgreSQL workloads interact with advanced analytics, caching layers, and AI/ML components, streamlining development and real-time data processing for modern applications.
🔗 https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-aurora-postgresql-integration-kiro-powers/
Amazon S3 Maximum Object Size Increased to 50 TB
AWS expanded the maximum supported object size in Amazon S3 to 50 TB , a significant increase that helps organizations store and process ultra-large datasets — such as high-resolution imaging, multimedia archives, and massive scientific data — without splitting data across multiple objects. This change simplifies data pipelines and reduces overhead for large file storage.
🔗 https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-s3-maximum-object-size-50-tb/
Amazon S3 Vectors Now Generally Available
Amazon S3 Vectors reached general availability, offering native support for vector storage and retrieval directly within S3. This capability makes it easier and more cost-efficient to build semantic search, recommendation systems, and other AI-powered applications at scale — without deploying separate vector databases.
🔗 https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-s3-vectors-generally-available/
Final Thoughts
At AWS re:Invent 2025 AWS announced some key features to optimize data workloads specially on the Cost side, allowing larger databases and data centric applications to have a better ROI and TCO on the cloud.
For engineering leaders shaping 2026 roadmaps, this is the perfect moment to step back and reassess your data strategy, what could be moved to the cloud now that having all your data in a central place has become more critical as we expand into AI and data analytics applications that will re-shape the future of the organizations and how customers and internal stakeholders interact with the vast data we have been accumulating for the past decade.
What do you think — do any of these announcements position you to better go to cloud and enable some of the new projects in your companies?
About MacondoTek
At MacondoTek, we help organizations modernize, scale, and innovate on AWS by combining deep cloud expertise with practical, real-world engineering experience. Our team specializes in serverless architectures, container platforms (ECS/EKS), AI-driven workloads, and end-to-end cloud modernization strategies tailored to each customer’s unique needs.
Whether you’re exploring how Lambda Managed Instances can reshape your compute strategy, planning an EKS/ECS modernization, or looking to accelerate your AI and data initiatives on AWS, our team can help you move faster with confidence.
If you’d like to learn more about our services, solutions, and past projects, we invite you to visit us at:
We’d love to connect, collaborate, and help your organization unlock more value from AWS.
References & re:Invent 2025 Announcements
The following official AWS announcements and summaries were referenced across this article:
- Top Announcements of AWS re:Invent 2025 https://aws.amazon.com/blogs/aws/top-announcements-of-aws-reinvent-2025/
- Database Savings Plans https://aws.amazon.com/about-aws/whats-new/2025/12/database-savings-plans-savings/
- AWS Database Features Launch Summary (re:Invent 2025) https://builder.aws.com/content/2y0CSKQ114Ivq2AHGeNbxW9TmlG/amazon-databases-reinvent-2025-features-launch-summary
- AWS Glue Zero-ETL for Self-Managed Databases https://aws.amazon.com/about-aws/whats-new/2025/12/aws-glue-zero-etl-self-managed-database-source/
- AWS DMS Schema Conversion: SAP Sybase ASE → PostgreSQL https://aws.amazon.com/about-aws/whats-new/2025/11/aws-dms-schema-conversion-sap-sybase-ase-postgresql/
- Amazon S3 Tables Enhancements (Intelligent-Tiering & Replication) https://aws.amazon.com/blogs/aws/announcing-replication-support-and-intelligent-tiering-for-amazon-s3-tables/
- Amazon RDS Updates and Licensing Changes https://aws.amazon.com/blogs/aws/category/database/amazon-rds/
These announcements collectively highlight AWS’s continued investment in cost optimization, database modernization, and AI-ready data platforms.