News For Dev-ops

Amazon SageMaker HyperPod now supports flexible instance groups, enabling customers to specify multiple instance types and multiple subnets within a single instance group. Customers running training and inference workloads on HyperPod often need to span multiple instance types and availability zones for capacity resilience, cost optimization, and subnet utilization, but previously had to create and manage a separate instance group for every instance type and availability zone combination, resulting in operational overhead across cluster configuration, scaling, patching, and monitoring. With flexible instance groups, you can define an ordered list of instance types using the new InstanceRequirements parameter and provide multiple subnets across availability zones in a single instance group. HyperPod provisions instances using the highest-priority type first and automatically falls back to lower-priority types when capacity is unavailable, eliminating the need for customers to manually...

Introduction: The Problem We recently conducted a customer engagement that started as a series of data science experiments and gradually moved into a production capability by "promoting" experimental code into MCP Tools or Agents. Throughout this article "tool" should be considered to be synonymous with Tool or Agent in the MCP sense. The production MCP Server was hosted in an Azure Container App and exposed a set of tools to GitHub Copilot in VS Code. We defined 22 requirements for this MCP Server. The relevant ones for this article are: The tools should be discoverable at runtime Individual tools must be enabled or disabled using Feature Flags The tools should be observable in the cloud The tools must be stateless - all state to be externalized Tools should be chainable, allowing one tool to invoke either in-process or external MCP tools. The Journey: Our Approach and Solution This section briefly describes the solution for each of the stated requirements...

AWS is announcing the general availability of Amazon EC2 C8in and C8ib instances powered by custom, sixth generation Intel Xeon Scalable processors, available only on AWS. These instances feature the latest sixth generation AWS Nitro cards. C8in and C8ib instances deliver up to 43% higher performance compared to previous generation C6in instances. C8in and C8ib instances deliver larger sizes and scale up to 384 vCPUs. C8in instances deliver 600 Gbps network bandwidth—the highest among enhanced networking EC2 instances—making them ideal for network-intensive workloads like distributed compute and large-scale data analytics. C8ib instances deliver up to 300 Gbps EBS bandwidth, the highest among non-accelerated compute instances, making them ideal for high-performance commercial databases and file systems. C8in instances are available in US East (N. Virginia), US West (Oregon), Asia Pacific (Tokyo), and Europe (Spain) regions. C8ib instances are available in US East (N. Virginia) and U...

Running Aspire apps from the CLI has a friction point: your terminal is locked to the AppHost process. While the app is running, you can't use the terminal to check logs or telemetry. You must open a new terminal window. Worse, if an agent has started your Aspire app to test a feature, the agent may decide to kill the app to free the terminal for a new command. Oops! Aspire 13.2 fixes these problems with detached mode , a way to run an Aspire app in the background. Aspire 13.2 also introduces a new suite of process management commands that give you real CLI control over your application lifecycle. Running in the Background The simplest feature is often the most useful. With detached mode, you start your AppHost and immediately get your terminal back: aspire start That's it. Your AppHost spins up in the background, and you're ready to run other commands in the same terminal. No more hunting for another shell window or tmux session. This is especially valuable during l...

A customer, via their customer liaison, reported quite some time ago that their program stopped working on Windows XP. (I told you it was quite some time ago.) The customer's investigations revealed that the problem occurred because their window was receiving message 0x0091 , and the parameters are wrong. Who is sending this message with the wrong parameters? Okay, first of all, how do you even know that the parameters are wrong? The message is not listed in winuser.h or in MSDN (as it was then called). We explained that message 0x0091 is an internal message that they should just pass to Def­Window­Proc unchanged. What makes the customer think that the message is being received with the wrong parameters? The customer said that their program was using that message as a custom message, and now, in addition to getting it when their program sends the message, they are also getting spurious copies of the message with WPARAM and LPARAM values that don't correspond to any val...

Amazon Bedrock, the platform for building AI applications and agents at production scale, now offers Claude Opus 4.7-- Anthropic's most capable Opus model to date -- delivering meaningful improvements across agentic coding, professional work, and long-running tasks for developers and enterprises building production AI applications.  Claude Opus 4.7 is an upgrade from Claude Opus 4.6, with stronger performance across the workflows teams run in production. Opus 4.7 works better through ambiguity, is more thorough in its problem solving, and folllows instructions more precisely. For coding, the model extends agentic capabilities with improved long-horizon autonomy, systems engineering, and complex code reasoning. For knowledge work, Claude Opus 4.7 advances professional tasks such as slides and document creation, financial analysis, and data visualization. For long-running tasks, the model stays on track...