Posts

Amazon SageMaker HyperPod now supports data capture for inference workloads - devamazonaws.blogspot.com

Amazon SageMaker HyperPod now supports data capture for inference workloads, enabling customers to record inference request and response payloads for model monitoring, compliance, debugging, and offline analysis. Organizations deploying generative AI and machine learning models on HyperPod need systematic visibility into the inputs flowing into their models and the outputs returned to clients to detect model drift, satisfy regulatory audit requirements, debug production issues, and build ground-truth datasets for fine-tuning. Previously, customers had to either accept limited operational visibility into their inference workloads or build expensive custom logging pipelines outside the HyperPod Inference Operator. With data capture, you can choose to record inference traffic at the SageMaker endpoint, at the load balancer, or at the model pod, depending on the level of visibility you need, and combine these options for layered observability. Captured data is delivered asynchronously to...

Amazon Managed Grafana now supports dual-stack connectivity (IPv6 and IPv4) - devamazonaws.blogspot.com

Amazon Managed Grafana now supports dual-stack connectivity, enabling workspaces to communicate over both Internet Protocol version 4 (IPv4) and Internet Protocol version 6 (IPv6). Dual-stack mode is available for workspaces running Grafana version 10.4 or later. With dual-stack support, customers can simplify their network stack by eliminating the need to manage overlapping address spaces in their VPCs. Customers migrating to IPv6 can connect to their Grafana workspaces over IPv6 while maintaining IPv4 compatibility, and those not yet on IPv6 can continue using IPv4-only connections. This is especially beneficial as the continued growth of the internet exhausts available IPv4 addresses.  Support for dual-stack connectivity on Amazon Managed Grafana is available in all  regions  where the service is generally available. To get started, update your workspace configuration via the Amazon Managed Grafana console, API, or ...

[MS] Just shows that nobody cares about debugging the parity flag any more - devamazonaws.blogspot.com

The x86-64 architecture inherited the parity flag (PF) from the x86-32, which in turn inherited it from the 8080, which inherited it from the 8008, which implemented it because it was the processor for the Datapoint 2200 serial terminal. The parity flag also has a secondary purpose of being a place for the FXAM (x87) and UCOMISD (SSE) instructions to record the results of floating point comparisons. You can still entice compilers into checking the parity flag by checking a value for NaN or performing a floating point equality or inequality comparison (because NaN always fails equality and inequality comparison). It turns out that the Windows debugging engine for x86-64 had a bug where it reported the parity flag as the opposite of what it actually is. When the parity flag was set, it said "po" instead of "pe", and vice versa. The fact that this went unreported for over two decades tells you that nobody cares about debugging the parity flag. A fix has gone in. ...

[MS] NuGet PackageReference for C++ Projects in Visual Studio - devamazonaws.blogspot.com

Native C++ projects in Visual Studio now support <PackageReference> , the modern, MSBuild-native way to declare NuGet package dependencies directly in your project file. This support is available experimentally for .vcxproj projects in the Visual Studio Insiders Channel starting with version 18.7. This feature has been the most upvoted feature request on Visual Studio Developer Community, and we're delivering it based on that feedback and in collaboration with other teams at Microsoft, including Windows and Azure. NuGet with PackageReferences can be useful for teams that develop both .NET and C++ projects (native or interop) that need a consistent way to deploy their binaries across their repos or to their consumers, or for managing dependencies that aren't C++ libraries, such as binary SDK packages. We continue to recommend vcpkg for acquiring and managing C++ libraries, as it is more specialized and flexible for these types of dependencies. What Is PackageReference...

Amazon Redshift adds ALTER TABLE for Iceberg tables and writes via the AWS Glue Data Catalog mount - devamazonaws.blogspot.com

Amazon Redshift now supports writing directly to Apache Iceberg tables via the AWS Glue Data Catalog (awsdatacatalog) mount and ALTER TABLE DDL statements to modify the schema, partitioning, and properties of Apache Iceberg tables. With write access through the auto-mounted awsdatacatalog, you can land Redshift transformations in your data lake for any engine to query without creating external schemas—particularly useful for Iceberg tables federated with AWS Lake Formation. Supported ALTER TABLE operations include ADD/DROP/ALTER columns, RENAME COLUMN, SET TABLE PROPERTIES to overwrite the default compression type, and ADD/DROP/REPLACE PARTITION FIELD to adapt partitioning strategies as data volumes grow. Previously, updating the structure of Iceberg tables required deleting the table and its data, adding complexity and latency to data pipelines. Tables modified by Redshift remain compatible with other Iceberg-compatible engines, including Amazon EMR and Amazon Athena , pres...

Amazon SageMaker Studio now supports GPU capacity reservation through SageMaker Flexible Training Plans - devamazonaws.blogspot.com

Amazon SageMaker Studio IDEs, including JupyterLab and Code Editor, now support GPU capacity reservations through SageMaker Flexible Training Plans (FTP) , giving you predictable access to high-demand, high-performance computational resources within your budget. By leveraging FTP, you can achieve up to 65% cost savings compared to On-Demand instances while running ML workflows in JupyterLab or Code Editor. FTP provides a fully self-serve procurement experience. To get started, navigate to the SageMaker FTP console and select your preferred instance type, reservation length, and start date for your Studio IDE workload. Review your order, complete the purchase, and wait for the plan to become active. When creating a Studio app from the SageMaker Studio UI, select your purchased plan from the Instance dropdown. SageMaker provisions the instance automatically with no infrastructure management required on your part. As your plan nears expiration, the IDE proactively notifies you, giving y...

AWS SAM CLI adds AWS CloudFormation Language Extensions support to accelerate local serverless development - devamazonaws.blogspot.com

AWS SAM CLI now supports AWS CloudFormation Language Extensions , enabling you to reduce duplication in your infrastructure as code (IaC) templates while retaining the full local development workflow. This accelerates your serverless development by letting you define resources once and iterate locally without waiting for cloud deployments. Developers frequently need to define multiple similar resources, such as Lambda functions, DynamoDB tables, or SNS topics, from a single template definition. However, developers who use SAM CLI to build, test, and deploy their serverless applications previously could not process templates that use CloudFormation Language Extensions. This required choosing between reducing template duplication and using SAM CLI for local development. Now, SAM CLI processes Language Extensions in memory for local operations while preserving your original template for CloudFormation deployment. You can define your resources once and test them locally across all SAM C...