InfoQ Homepage Architecture & Design Content on InfoQ
-
Google Cloud Launches A4 VMs with NVIDIA Blackwell GPUs for AI Workloads
Google Cloud has launched A4 VMs, powered by NVIDIA's Blackwell B200 GPUs, revolutionizing AI workloads with a 2.25x performance boost and advanced networking. Key features include tight integration with Google Kubernetes Engine and Vertex AI, enhancing deployment and management of large-scale projects. This innovative partnership sets a new standard in cloud infrastructure for AI.
-
JavaOne 2025 Day 1: Our World, Moved by Java
JavaOne 2025, celebrating the 30th birthday of the Java programming language, was held at the Oracle Conference Center in Redwood Shores, California. This three-day event consisted of 80 sessions and hand-on labs, and two keynotes. Day One, scheduled on March 18th, 2025, included the opening keynote and presentations from various conference tracks.
-
Dapr Agents: Scalable AI Workflows with LLMs, Kubernetes & Multi-Agent Coordination
Introducing Dapr Agents—a groundbreaking framework for creating scalable AI agents using Large Language Models (LLMs). With robust workflows, multi-agent coordination, and cloud-neutral architecture, it enables enterprises to deploy thousands of resilient agents. Built on Dapr’s proven infrastructure, Dapr Agents ensures reliability and observability in AI-driven applications.
-
AWS AppSync Events Adds Direct WebSocket Message Publishing
AWS has boosted AppSync Events with direct WebSocket message publishing, enhancing real-time communication. This serverless API facilitates seamless event broadcasting to numerous subscribers and simplifies development by allowing combined publishing and receiving via WebSocket. Discover a faster way to build engaging applications today!
-
Cloudflare Launches Media Transformations: Optimizing Short-Form Video
Cloudflare’s Media Transformations simplifies video optimization for short-form content, eliminating complex pipelines and allowing users to enhance videos directly from their existing storage. With URL-based parameters for automation, it offers features like format conversion and frame extraction. Currently in beta, it's free until Q3 2025 and aims to streamline workflows across platforms.
-
Java 24 Delivers New Experimental and Many Final Features
Oracle has released version 24 of the Java programming language and virtual machine. As the third non-LTS release since JDK 21, the final feature set includes 24 JEPs. Two of these - Generational Shenandoah and Compact Object Headers - are new experimental features. Also, nearly half of these features are final, some of which have gone through the incubation and preview processes.
-
Azure Database for MySQL Trigger for Azure Functions in Public Preview
Microsoft's Azure Database for MySQL now features a public preview of triggers for Azure Functions, enabling real-time data processing with event-driven automation. This innovative integration allows developers to track table changes seamlessly, enhancing analytics, compliance, and security, while scaling efficiently within Azure's serverless framework.
-
Meta Unifies Facebook’s Video Delivery System across Mobile and Web Apps
Meta finalized efforts to consolidate Facebook’s video delivery system by migrating video experiences from older Watch product to more recent Reels product, which became the basis of the unified system. The unification process required changes across mobile UI, server backend, and ranking systems while ensuring a seamless transition for billions of users.
-
Google Cloud's AI Protection: a Solution to Securing AI Assets
Google Cloud introduces AI Protection, a solution to safeguard against generative AI threats. Managing AI risks through vulnerability assessments, security policies, and proactive threat management enhances asset protection. Integrating with Google’s Security Command Center offers a centralized view of IT posture and advanced security intelligence for robust AI system defense.
-
Learnings from Working with Programming Rules and Guidelines
Programming rules and guidelines improve code consistency, but misapplication can lead to poor results. Arne Mertz suggests that software developers selectively adopt rules and guidelines, and document deviations with clear explanations. They can discuss their experiences in communities or during their daily work, to foster collaboration and improve code quality without unnecessary bureaucracy.
-
Microsoft Enhances Azure Elastic SAN with Auto Scale, Snapshot Support, and CRC Protection
Microsoft's Azure Elastic SAN, launched in early 2024, revolutionizes cloud block storage with unique autoscale capabilities, snapshot support, and CRC protection for enhanced data integrity. This fully managed solution simplifies storage management and optimizes costs, making it ideal for businesses seeking efficient, high-availability solutions in the cloud.
-
Local Development with Workflow Studio for Step Functions
AWS has enhanced its Workflow Studio for Step Functions, now integrated into Visual Studio Code via the AWS Toolkit. This allows developers to create and edit state machines locally with intuitive visual tools. Key features include Design and Code modes, localized testing capabilities, and support for ASL definitions, streamlining the development of distributed applications and workflows.
-
New Programming Language Vine Based on Interaction Nets
Introducing Vine, an innovative programming language built on interaction nets that seamlessly integrates functional and imperative paradigms. Statically typed and compiled, Vine's intuitive syntax empowers developers to explore parallel computing's potential. With a robust compiler and active community, Vine invites collaboration and creativity.
-
Java News Roundup: Milestone Releases of Spring Cloud, GlassFish and Grails, Devnexus 2025
This week's Java roundup for March 3rd, 2025, features news highlighting: milestone releases of Spring Cloud 2025.0.0, GlassFish 8.0.0 and Grails 7.0.0; point releases of Spring gRPC 0.4.0, Helidon 4.2.0, Quarkus 3.19.2 and JHipster 1.29.1 and 1.29.0; the fourth release candidate of Netty 4.2.0; and Devnexus 2025.
-
Vercel Fluid: a New Compute Model and an Alternative to Serverless?
Vercel has recently introduced Vercel Fluid, an elastic compute model that allows a single worker to handle multiple requests, similar to a traditional server, while preserving the elasticity of serverless. By scaling functions before instances, Fluid maximizes available compute time, optimizing compute footprint and resource efficiency for long-running tasks and AI inference.