News For Dev-ops

AWS IoT Core, AWS IoT Device Management, AWS IoT Device Defender and AWS IoT Greengrass are now available in Europe (Spain) and Asia Pacific (Malaysia) Regions. This comprehensive suite of IoT services help customers securely connect, monitor, troubleshoot, and manage IoT devices at scale. With the region expansion update, organizations operating in these regions can better serve their local customers and unlock multiple benefits, including faster response times, stronger data residency controls, and reduced data transfer expenses. AWS IoT Core helps connect, manage, and secure billions of IoT devices. It routes trillions of messages to other IoT devices and AWS endpoints, through bi-directional industry standard protocols, such as MQTT, HTTPS, and LoRaWAN. AWS IoT Device Management provides the flexibility to remotely monitor, manage, and update IoT device fleet at scale while AWS IoT Device Defender empowers you to monitor security posture and detect anomalies. AWS IoT Greengrass i...

Amazon Simple Notification Service (Amazon SNS) now supports message group IDs in standard topics, enabling fair queue functionality for all subscribed Amazon SQS standard queues. This feature allows you to mitigate noisy neighbor impact in all multi-tenant standard queues subscribed to a SNS standard topic by ensuring that high-volume or slow-processing messages from one tenant don't delay messages from other tenants. When you include a message group ID in messages sent to your Amazon SNS standard topic, the topic automatically forwards these IDs to all subscribed Amazon SQS standard queues, activating fair queue behavior across those queues. This capability is particularly valuable for SaaS applications that use Amazon SNS to distribute messages to multiple processing queues, event-driven architectures serving multiple customers, and microservices that need to maintain quality of service across different request types. This feature is now available in all AWS commercial and ...

AWS IoT Core now supports message queuing for MQTT shared subscriptions, helping to enable IoT applications to maintain message delivery reliability during network disruptions. When all the members of a shared subscription group are disconnected or cannot process the incoming message volume, the service automatically queues Quality of Service 1 (QoS1) messages on persistent sessions. Queued messages are automatically delivered when shared subscribers reconnect or when additional shared subscribers are added to the group. This process helps preserve critical data during connectivity challenges and peak loads.. The feature is fully compatible with both MQTT v3.1.1 and 5.0 clients, and is now available in all AWS Regions where AWS IoT Core is supported. Queued messages will be billed at the standard messaging rate; visit the pricing page for more information about message pricing and get started with the AWS IoT Core documentation for implementation details. Post Updated on July 31,...

We’re excited to introduce updates to GitHub Copilot for Xcode, bringing new ways to interact with AI and streamline your development workflow. With Copilot Vision, you can upload and chat with images—like screenshots and UI designs—directly in Xcode. We’ve also introduced global and workspace-level Custom Instructions, locale-aware responses, and a new auto-attach chat window to make interactions smoother and more personalized. What’s New Copilot Vision Support You can now upload images—such as screenshots, diagrams, or UI mockups—directly into Copilot Chat. Whether it’s a screenshot of an error, a diagram, or a UI mockup, Copilot Vision helps you understand and reason about visual content right inside Xcode. To help you get started, here is a quick demo—building a classic Snake game in Swift, right inside Xcode. [video mp4="https://devblogs.microsoft.com/xcode/wp-content/uploads/sites/87/2025/07/All-HandsXcodeCopilotVisionDemoVoiceS1.mp4"][/video] Custom Instructions ...

Amazon Aurora MySQL-Compatible Edition now supports a maximum storage limit of 256 TiB, doubling the previous limit of 128 TiB. This enhancement allows customers to store and manage even larger datasets within a single Aurora database cluster simplifying data management for large-scale applications and supporting the growing data needs of modern applications. Customers only pay for the storage they use, with no need for upfront provisioning of the full 256 TiB. To access the increased storage limit, upgrade your cluster to supported database versions. Once upgraded, Aurora storage will automatically scale up to 256 TiB capacity based on the amount of data in the cluster volume. Visit  technical documentation  to learn more about supported versions. This new storage volume capacity is available in all  AWS regions  where Aurora MySQL and Aurora PostgreSQL is available.  Amazon Aurora is designed for unparalleled high performance and availability at global sca...

Starting today, Amazon Aurora MySQL - Compatible Edition 3 (with MySQL 8.0 compatibility) will support MySQL 8.0.42 through Aurora MySQL v3.10. In addition to several security enhancements and bug fixes, MySQL 8.0.42 contains performance improvements for parallel replication using writeset dependency tracking , as well as enhanced debugging capabilities within the InnoDB storage engine. Aurora MySQL 3.10 includes an increase in maximum storage capacity from 128 TiB to 256 TiB , allowing customers to manage larger database workloads within a single database cluster. Aurora MySQL 3.10 also introduces in-memory relay log optimization that improves binary log replication performance by caching relay log content in memory, reducing commit latency and minimizing storage I/O operations on binlog replicas. For more details, refer to the Aurora MySQL 3.10 and MySQL 8.0.42 release notes. To upgrade to Aurora MySQL 3.10, you can initiate a minor version upgrade manually by modifying your DB...

A while ago, I investigated ways of awaiting multiple C++/WinRT corotuines and capturing the results . I started with winrt:: when_all() , but really I could have started even earlier. The easy way to await multiple coroutines is to await each one! auto op1 = DoSomething1Async(); auto op2 = DoSomething2Async(); auto result1 = co_await op1; auto result2 = co_await op2; This starts two operations, and only after they have started do we wait for them in sequence. This works great, as long as you understand that if an exception occurs in co_await op1 , then the second operation op2 just keeps on running and eventually has nobody to report results to. This is probably okay for the most part: The second operation just keeps on running pointlessly. Not necessarily the most efficient, but probably not harmful. If you want to cancel the second operation if the first operation fails, you can do that: auto op1 = DoSomething1Async(); auto op2 = DoSomething2Async(); auto op2cancel = wil::...

Amazon Relational Database Service (Amazon RDS) for Oracle has expanded its instance offerings in AWS GovCloud (US) Regions. The M7i and R7i instances, which support a maximum instance size of 48xlarge, are Intel-based offerings delivering 50% more vCPU and memory compared to M6i and R6i instance types. The X2idn instances are optimized for memory-intensive workloads with up to 2TiB of memory. M7i, R7i and X2idn instances are available for Amazon RDS for Oracle in Bring Your Own License model for both Oracle Database Enterprise Edition (EE) and Oracle Database Standard Edition 2 (SE2). You can launch the new instance in the Amazon RDS Management Console or using the AWS CLI or SDK. Refer to the Amazon RDS for Oracle Pricing page to see pricing details and regional availability. Post Updated on July 30, 2025 at 10:53PM

We’re excited to announce the General Availability (GA) of two long-standing capabilities that address critical needs for SQL developers and enterprise customers in Azure SQL Database and Azure SQL Managed Instance, configured with the Always-up-to-date update policy: UNISTR function for Unicode character representation ANSI SQL string concatenation operator ( || ) for standard-compliant string operations These additions improve ANSI SQL compliance in Azure SQL, streamline migration from other platforms, and boost developer productivity. UNISTR function: Unicode made easy The UNISTR function provides support for Unicode string literals by letting you specify the Unicode encoding value of characters in the string, making it especially useful for working with international or special characters. If you want to include characters like emojis, accented letters, or symbols from non-Latin scripts, UNISTR lets you encode them directly using their Unicode values. Returns Unic...

Ever wonder why your agent's answers can be right but for the wrong reasons? Let's dive into why that matters, with a brief foray into epistemology (bear with me!) Knowledge: More Than Just Right Answers Imagine asking someone the time, and they confidently reply, “2:30 PM.” They're correct—but what if their watch stopped exactly 12 hours ago, and they just happened to get lucky? This illustrates a classic philosophical point: the difference between a true belief and a justified true belief . And no, this isn’t just a theoretical curiosity (yes, philosophy folks are sensitive about that critique). In practice, we do care about how people form their beliefs—and, by analogy, how AI agents arrive at their responses. In LLM system design, we evaluate not only if an agent is correct, but how it got there. A key metric here is tool correctness —a measure of whether the AI used the right tools, in the right way, to arrive at its conclusion. This is where the analogy to ju...

Join Robert Green and Leslie Richardson for a full day of live sessions, demos, and interviews. Live from Microsoft HQ – Stream Tuesday’s Sessions with Visual Studio Toolbox! On Tuesday, August 5 , join us for a special edition of Visual Studio Toolbox Live —broadcast from VS LIVE! Redmond and hosted by Leslie Richardson and Robert Green . You’ll get a full day of Microsoft-led sessions, live studio commentary, behind-the-scenes interviews, and exclusive insights straight from campus. 👉 Click here to set a YouTube reminder so you don’t miss a moment. 🎤 What You’ll See on Tuesday We’re streaming five sessions led by Microsoft product teams across Visual Studio, Azure, GitHub Copilot, and .NET. Whether you're looking to boost your daily productivity or build your next-gen AI app, this day is packed with learning. 🛠 8:00 AM – Getting the Most out of the Latest in Visual Studio Harshada Hole & Jessie Houghton, Microsoft See the newest productivity boosts, AI-assisted deb...

Amazon Bedrock Data Automation (BDA) now supports DOC/DOCX files for document processing and H.265 encoded video files for video processing. This update allows you to build more comprehensive and efficient multimodal data analysis pipelines by working directly with these popular formats. With support for DOC/DOCX, you can now seamlessly process your Word documents, reports, and other text-based content without the intermediate step of converting them to PDF. This streamlines your workflows, allowing for faster text extraction and analysis. H.265 support enables you to process high-quality video files more efficiently. The H.265 codec provides superior video quality at significantly smaller file sizes compared to older codecs, speeding up processing times for your video analysis workloads. Amazon Bedrock Data Automation is available in 7 AWS regions: Europe (Frankfurt), Europe (London), Europe (Ireland), Asia Pacific (Mumbai, Asia Pacific (Sydney), US West (Oregon) and US East (N....

Today, AWS announces the general availability of multivariate anomaly detection in AWS IoT SiteWise. This native capability enables industrial customers to automatically identify abnormalities across their equipment and asset data. This capability simplifies the implementation of predictive and preventative maintenance strategies without requiring machine learning expertise, helping to improve operational efficiency and reduce downtime. Multivariate anomaly detection in AWS IoT SiteWise helps to address a critical challenge faced by industrial operations: the need to prevent costly downtime and maintain peak efficiency. By embedding this capability directly within AWS IoT SiteWise, industrial customers can now detect equipment anomalies across their entire fleet of assets without writing code or managing complex infrastructure, lowering the barrier to entry for implementing effective predictive maintenance strategies. This feature is particularly effective for assets like turbines, c...

A recurring problem I encounter when reviewing API proposals is that teams tend not to be precise in their use of terminology. This casualness is inevitable when you work with a feature for a long time and develop notational shortcuts, but the people who are learning your API don't have the same level of familiarity that you do, and shifting terminology tends to create confusion. For example, there was an API proposal that included two methods. runtimeclass Widget { void EnableFilter(WidgetFilter filter); Boolean AreAnyFiltersApplied(); } The first method talks about "enabling" but the second talks about "applying". For somebody encountering this API for the first time, the existence of two different terms raises questions. Is enabling a filter the same as applying it? Or are there two steps to making a filter active, first you enable it, then you apply it? (Or do you apply it first, and then enable it?) When I asked the team, they said that enablin...

The Amazon Connect Contact Control Panel (CCP) now features an updated user interface using Cloudscape Design System components to enhance agent productivity and focus. The refreshed user interface includes a updated visual style for colors and buttons, and more visually consistent UI elements across interfaces. This visual refresh provides a more intuitive and streamlined experience for your contact center agents while maintaining the familiar layout and functionality to minimize impact and training requirements. This updated user interface is available in all AWS regions where Amazon Connect is offered. To learn more, please see the documentation . Post Updated on July 28, 2025 at 05:33PM

The Model Context Protocol (MCP) has quickly established itself as the go-to standard for seamlessly connecting AI agents with a variety of external tools and services. In just a few months, it has become a prominent topic at developer events and forums, gaining adoption from companies of all sizes. To meet the rapid pace of advancements, we ran a series of beginner-friendly training events on MCP. If you didn’t catch these events live, don’t worry! They’re all linked below for you to watch at your own pace. In these training sessions, our presenters ran through... What is MCP? How to use an MCP server in your applications. And how to build an MCP server. Sessions In total, we ran the training 18 times to reach developers using different programming languages. 🌍 Language ⚙️ Technology 🎥 Watch 💻 GitHub Repo English #️⃣ C# 📺 Watch Video 📚 View Code English ☕ Java 📺 Watch Video 📚 View Code English 🐍 Python 📺 Watch Video 📚 View Code English #️⃣ C# ...

Multi-tenant applications face a fundamental challenge: how to efficiently store and query data for tenants of vastly different sizes? Consider the typical scenario where your platform serves both enterprise clients with hundreds of thousands of users, as well as small businesses with just a handful. With traditional database partitioning strategies you are likely to run into these common issues: Partition imbalance : Large tenants create oversized partitions while small tenants waste allocated resources Hot partitions : High-activity tenants overwhelm individual database partitions, creating performance bottlenecks Inefficient queries : User-specific lookups require scanning entire tenant datasets Resource contention : Mixed workloads compete for the same database resources Azure Cosmos DB has been a go-to solution for  multi-tenant applications  due to its global distribution, automatic scaling, and flexible data models. Its partition-based architecture naturally align...

Amazon Connect now supports AWS CloudFormation for Outbound Campaign message template attachments, enabling you to create, manage, and deploy these template attachments using AWS CloudFormation. This enhancement allows customers to define and deploy attachments, such as images or documents, as part of the existing MessageTemplate CloudFormation resource. Message template attachments are used in outbound email campaigns to enrich content and improve engagement. With this launch, customers can now manage attachments programmatically through infrastructure as code, ensuring consistency, repeatability, and automation across staging, test, and production environments. This feature is available in all AWS Regions where Amazon Connect Outbound Campaigns are supported. For more information, see the AWS Region table. To learn more, see Message templates in the Amazon Connect Administrator Guide or visit the Amazon Connect product page. Post Updated on July 25, 2025 at 08:20PM

In today’s fast-paced software development cycles, manual testing often becomes a significant bottleneck. Our team was facing a growing backlog of test cases that required repetitive manual execution—running the entire test suite every sprint. This consumed valuable time that could be better spent on exploratory testing and higher-value tasks. We set out to solve this by leveraging Azure DevOps’ new MCP server integration with GitHub Copilot to automatically generate and run end-to-end tests using Playwright . This powerful combination has transformed our testing process: Faster test creation with AI-assisted code generation Broader test coverage across critical user flows Seamless CI/CD integration , allowing hundreds of tests to run automatically On-demand test execution directly from the Azure Test Plans experience (associating Playwright JS/TS tests with manual test cases is coming soon. Keep an eye on our release notes for the announcement.). By automating our testing...

Amazon Connect now supports AWS CloudFormation for quick responses, enabling customers to deploy and manage quick responses using AWS CloudFormation templates. Quick responses allow contact center agents to access pre-configured messages to respond consistently and efficiently to common customer inquiries. With AWS CloudFormation, administrators can now define and deploy these quick responses across environments in a scalable and repeatable way. Using AWS CloudFormation, organizations can standardize agent communications and reduce manual configuration between Amazon Connect instances. For example, you can use AWS CloudFormation templates to roll out updated response sets for seasonal campaigns or regulatory compliance across multiple Connect environments. This launch simplifies deployment and integrates seamlessly with continuous delivery pipelines. This feature is available in all AWS Regions where Amazon Connect is offered. For a full list of supported Regions, see the AWS Regio...

Amazon Relational Database Service (Amazon RDS) for PostgreSQL zero-ETL integration with Amazon Redshift is now generally available, enabling near real-time analytics and machine learning (ML) on petabytes of transactional data. With this launch, you can create multiple zero-ETL integrations from a single Amazon RDS database, and you can apply data filtering for each integration to include or exclude specific databases and tables, tailoring replication to your needs. You can also use AWS CloudFormation to automate the configuration and deployment of resources needed for zero-ETL integration. Zero-ETL integrations make it simpler to analyze data from Amazon RDS to Amazon Redshift by removing the need for you to build and manage complex data pipelines and helping you derive holistic insights across many applications. Within seconds of data being written to Amazon RDS for PostgreSQL, the data is replicated to Amazon Redshift. Using zero-ETL, you can enhance data analysis on near real-tim...

Starting today, Amazon EC2 Instance Connect and EC2 Serial console are also available in AWS Asia Pacific (Malaysia), AWS Asia Pacific (Thailand) and AWS Mexico (Central) regions. EC2 Instance Connect allows customers to connect to their instances over Secure Shell (SSH) with a single click from the EC2 console and single command from the AWS CLI. Customers can manage access to instances with AWS Identity and Access Management (IAM) policies as well as generate one-time use SSH keys enhancing the security posture for connectivity. EC2 Serial Console provides customers a simple way to interactively troubleshoot boot and network connectivity issues by establishing a connection to the serial port of an instance. EC2 Serial Console is ideal for situations where you are unable to connect to your instance via normal SSH or RDP connections. To get started, you can navigate to the Amazon EC2 console , select the instance you want to connect to and with a single click connect using EC2 Seri...

Starting today, Amazon Elastic Compute Cloud (Amazon EC2) P6-B200 instances accelerated by NVIDIA Blackwell GPUs are available in US East (N. Virginia) region. These instances offer up to 2x performance compared to P5en instances for AI training and inference. P6-B200 instances feature 8 Blackwell GPUs with 1440 GB of high-bandwidth GPU memory and a 60% increase in GPU memory bandwidth compared to P5en, 5th Generation Intel Xeon processors (Emerald Rapids), and up to 3.2 terabits per second of Elastic Fabric Adapter (EFAv4) networking. P6-B200 instances are powered by the AWS Nitro System, so you can reliably and securely scale AI workloads within Amazon EC2 UltraClusters to tens of thousands of GPUs. Post Updated on July 23, 2025 at 08:42PM

MCP has evolved significantly beyond its original goal of "providing context to LLMs." With recent enhancements including resumable streams , elicitation , sampling , and notifications ( progress and resources ), MCP now provides a robust foundation for building complex agent-to-agent communication systems. In this article, you'll learn: How to build agent-to-agent communication with MCP capabilities  where MCP hosts and tools both act as intelligent agents Four key capabilities that make MCP tools "agentic" - streaming, resumability, durability, and multi-turn interactions How to build long-running agents with a complete Python implementation (an MCP server with a travel agent and research agents that stream updates, request input and are resumable.) [caption id="attachment_19561" align="alignnone" width="2500"] The MCP spec now supports agentic capabilities – specifically, tools are now resumable, can stream progress upd...