Postgres Archives

Tech Optimizer

April 27, 2026

Why developers are betting on Postgres for AI

Organizations are intensifying the development of AI applications and agents, which rely on access to existing enterprise data to avoid inaccuracies in responses, a phenomenon known as "hallucination." Jensen Huang, CEO of Nvidia, and Phillip Merrick, co-founder of pgEdge, emphasize that structured data is essential for AI effectiveness. PostgreSQL is highlighted as a preferred database for AI applications, with 66% of respondents in the Stack Overflow 2025 Developer Survey indicating they wish to continue using it. Merrick notes PostgreSQL's advantages, including its open-source model, scalability, and ability to handle both structured and unstructured data. The pgEdge toolkit supports the entire lifecycle of AI application development, including document ingestion and vector embedding generation. PostgreSQL offers deployment flexibility and robust security features, making it suitable for critical sectors like finance and healthcare.

Tech Optimizer

April 23, 2026

Take Control: Customer-Managed Keys for Lakebase Postgres

Encryption at rest is essential for cloud environments, especially in regulated sectors. Lakebase offers Customer Managed Keys (CMK), allowing organizations to control their encryption keys from preferred Key Management Services (KMS) like AWS KMS, Azure Key Vault, or Google Cloud KMS. Lakebase CMK manages both persistent storage and ephemeral compute, ensuring data security. Lakebase's architecture separates storage and compute layers, enabling elastic scaling and serverless operations. The storage layer maintains persistent data, while the compute layer consists of dynamic Postgres instances. This separation poses encryption challenges, addressed by Lakebase CMK through a hierarchical Envelope Encryption model. The key hierarchy includes: 1. Customer Managed Key (CMK): The Root of Trust in the cloud KMS, never accessed in plaintext by Databricks. 2. Key Encryption Key (KEK): A transient key for wrapping data keys. 3. Data Encryption Keys (DEKs): Unique keys for each data segment, stored in an encrypted state. When data access is needed, Lakebase retrieves DEKs by unwrapping keys from the KMS. If a key is revoked, access to the data is denied, and ephemeral compute instances are terminated. In practice, CMK applies to: 1. Persistence Layer (Storage): All data segments are encrypted with keys controlled by the CMK. 2. Ephemeral Layer (Compute): Each compute instance generates a unique ephemeral key, and upon revocation, the instance is terminated, destroying in-memory keys. CMK implementation follows a delegation model, allowing Security Admins to manage keys without accessing data. The process includes key configuration, workspace binding, and lifecycle management. Key rotation is seamless, requiring no data re-encryption. All cryptographic operations involving the CMK are logged by the cloud provider's audit service. Lakebase CMK is available for Enterprise tier customers, allowing organizations to manage their encryption keys effectively.

Tech Optimizer

April 22, 2026

pgEdge Launches AI DBA Workbench, an AI Co-Pilot for Database Administrators

pgEdge has introduced the AI DBA Workbench for PostgreSQL, an AI-driven tool designed to assist teams in managing PostgreSQL environments. The tool addresses the talent gap in database management, as the demand for experienced database administrators exceeds supply. The AI DBA Workbench continuously monitors PostgreSQL performance metrics and employs a three-tier anomaly detection system to identify potential issues. It includes an integrated assistant named Ellie, which provides diagnostic workflows and SQL code recommendations. The Workbench is compatible with PostgreSQL version 14 and above, is open source, and available for public download on GitHub. It is also included in pgEdge Enterprise Postgres, which offers enterprise-grade solutions. Support for the Workbench is available for pgEdge Enterprise Postgres subscribers.

Tech Optimizer

April 20, 2026

Databricks Postgres Gets Customer Key Control

Databricks has introduced customer-managed keys (CMK) for its Lakehouse Postgres offering, allowing users to control their data encryption and utilize their own Key Management Service (KMS) from major cloud providers like AWS KMS, Azure Key Vault, or Google Cloud KMS. The Lakehouse Postgres architecture features a hierarchical envelope encryption model with three tiers: the customer-managed root key (CMK), a Key Encryption Key (KEK) used by Databricks' Key Manager Service, and unique Data Encryption Keys (DEKs) for each data segment. If a CMK is revoked, access to the data is denied, and active compute instances are terminated, serving as a failsafe for high-compliance workloads. The implementation allows for granular control over key management and supports seamless key rotation without data re-encryption or downtime. All cryptographic operations are logged within the customer's cloud audit services, available to Databricks Enterprise tier customers.

Tech Optimizer

April 17, 2026

Why Postgres wants NVMe on the hot path, and S3 everywhere else

Efforts to merge storage roles into a single solution are ongoing, particularly with Amazon S3's durability and cost-effectiveness. In PostgreSQL, achieving a durable commit requires flushing the Write-Ahead Log (WAL) before signaling transaction completion, which can take tens of microseconds on high-performance NVMe drives but extend to milliseconds on slower storage. This latency impacts Online Transaction Processing (OLTP) systems and user response times. Benchmark studies show that systems with faster local storage outperform those with slower alternatives as workloads exceed memory capacity. The fsync operation in PostgreSQL is a commitment rather than a simple write, with enterprise-grade SSDs performing better due to power-loss protection. Read operations also face challenges, as PostgreSQL's need for small, latency-sensitive reads conflicts with S3's design for larger, higher-latency requests. As the working set exceeds memory, storage latency becomes a critical performance factor. Modern managed PostgreSQL systems typically do not place object storage in the critical commit path, instead maintaining a fast log or cache close to the database while relegating colder data to remote storage. Recent PostgreSQL developments, such as asynchronous I/O support in version 18, aim to leverage fast storage more effectively. S3 is valuable for tasks like WAL archiving and backups, but these should be kept separate from the commit path to avoid resource contention. The solution involves using both NVMe and S3, with fast storage managing commits and cache misses, while object storage handles archives and backups. PostgreSQL performs best when hot and cold storage functions are clearly delineated.

Tech Optimizer

April 17, 2026

pg_textsearch 1.0: Cómo hemos construido un motor de búsqueda BM25 en las páginas de Postgres

Recent advancements in artificial intelligence (AI) and robotics are reshaping industries and enhancing everyday life. The integration of AI with robotics has led to the development of intelligent machines capable of performing complex tasks autonomously. Key applications include: - **Manufacturing**: AI-driven robots streamline production lines, reduce downtime, and improve quality control through real-time data analysis for predictive maintenance. - **Healthcare**: Robots assist surgeons with precision tasks and support patient care, while AI algorithms enhance the speed and accuracy of diagnoses. - **Logistics**: Autonomous delivery drones and self-driving vehicles are transforming supply chain management by significantly reducing delivery times and operational costs. Machine learning has enabled robots to move beyond pre-programmed instructions, allowing them to engage in decision-making, analyze large data sets, and exhibit emotional intelligence. Ethical considerations regarding the responsible design and implementation of these technologies are becoming increasingly important to address potential biases and maintain public trust.

Tech Optimizer

April 17, 2026

EDB Spearheads the Year of the Agentic Workforce with Industry Recognition, Ecosystem Momentum, and Continued Postgres® Leadership

EnterpriseDB (EDB) announced that its product, EDB Postgres AI (EDB PG AI), won the "Data Management Innovation of the Year" at the 2026 Data Breakthrough Awards. EDB PG AI is recognized for transforming Postgres into a unified data and AI platform, facilitating hybrid management and observability across workloads. EDB contributes 30% to the open-source Postgres project, leading contributions among commercial entities. EDB PG AI was highlighted at the NVIDIA GTC 2026 and received multiple accolades, including recognition in CRN's 2026 AI 100 and Sustainability Innovation of the Year. EDB hosts a podcast, AI & Data Horizons, featuring industry experts discussing responsible AI use and data sovereignty. EDB PG AI is described as the first open, enterprise-grade sovereign data and AI platform, capable of unifying various workloads while ensuring compliance and security.

Tech Optimizer

April 16, 2026

Deploy Postgres and MySQL databases with PlanetScale + Workers

Cloudflare has deepened its partnership with PlanetScale to enhance the developer experience by providing Cloudflare Workers with access to Postgres and MySQL databases. Users will be able to create PlanetScale databases directly from the Cloudflare dashboard and API, with billing handled through their Cloudflare accounts. Cloudflare credits can be applied towards PlanetScale databases. The integration allows developers to link their PlanetScale accounts and create Postgres databases from the Cloudflare dashboard. Starting next month, new PlanetScale databases will be billed directly to Cloudflare accounts. The integration uses Hyperdrive for database connectivity, managing connection pools and query caching. PlanetScale offers performance and reliability, allowing developers to choose between Postgres and Vitess MySQL. Cloudflare users can deploy PlanetScale databases directly from Cloudflare, with features like query insights and usage breakdowns included. A single node on PlanetScale Postgres starts at a specified monthly rate. Developers can configure Workers to minimize latency by placing them near the primary database, with future plans to automate this process. Currently, billing for existing PlanetScale databases is processed through PlanetScale, but new databases will be billed through Cloudflare starting next month. Further enhancements and API integration are planned, with community feedback encouraged.

Tech Optimizer

April 15, 2026

Postgres to Iceberg in 13 minutes: How Supermetal compares to Flink, Kafka Connect, and Spark

Supermetal introduced Iceberg sink support, allowing for data transfer from Postgres to Iceberg. In a performance comparison, Supermetal completed snapshotting in 13 minutes, while Flink took 90 to 116 minutes, Kafka Connect required 120 minutes, and Spark exceeded 3 hours. The tests focused on throughput during the snapshotting phase rather than latency. The setup used the TPC-H dataset with a scale factor of 50 and involved AWS infrastructure. Supermetal operates independently of Kafka, enabling direct data delivery from source to sink. It utilizes a JSON configuration file for deployment, supporting various catalogs and advanced configuration options. Supermetal's Iceberg writer can adjust settings based on the CDC source phase. Throughout the tests, Supermetal maintained low CPU and memory usage, while other tools faced challenges with performance and resource management. Data synchronization was validated across all tools, with Supermetal achieving at least 7x faster snapshotting performance without tuning.

Tech Optimizer

April 11, 2026

Overcoming barriers to database modernisation in the AI age

Enterprises in Malaysia are transitioning from legacy systems to modern infrastructure to facilitate AI deployment. A roundtable discussion highlighted the challenges of AI integration, emphasizing the need to reduce costs associated with outdated systems. Organizations are adopting hybrid cloud approaches and utilizing various databases to manage extensive data across multiple applications. The push for AI is driven by management and customer expectations, but employee willingness to upskill remains a challenge. Not all challenges require AI solutions, and starting with smaller use cases can lead to successful scaling. The adoption of open-source database systems like Postgres is increasing, necessitating reliable support to address issues and ensure application availability. Data sovereignty is a concern for enterprises operating in mixed environments, and EDB Postgres AI offers a platform that combines security with cloud agility. Reducing infrastructure costs is essential for freeing up resources for new initiatives.