Amazon Athena unveils hourly reservations

Amazon Athena unveils hourly reservations

Updates related to Synapse Link, Azure Database, CodeCatalyst, Kinesis, Athena, Aurora, Dataflow and more

NPW Research

App Building

Azure Synapse Link is now compatible with Azure Cosmos DB's continuous backup for near-real-time analytics. UPDATE

Flexible Server adds auto-grow

Azure Database for PostgreSQL Flexible Server now features storage auto-grow and online disk scaling. UPDATE

New roles in Amazon CodeCatalyst

Amazon CodeCatalyst introduces four new user roles for enhanced project and space access control. UPDATE

Amazon Athena unveils hourly reservations

Amazon Athena introduces 1-hour Provisioned Capacity reservations, allowing flexible, cost-optimized workload management. UPDATE

Kinesis doubles On-Demand throughput

Amazon Kinesis Data Streams now supports a doubled On-Demand write throughput limit of 2 GB/s. UPDATE

RDS enhances Blue/Green Deployments

Amazon RDS's Blue/Green Deployments now support Amazon Aurora and RDS PostgreSQL, streamlining updates. UPDATE

Amazon Aurora MySQL speeds restarts

New optimizations in Amazon Aurora MySQL reduce database restart times by up to 65%. UPDATE

AI/ML

Dataflow enhances AI, developer tools, and efficiency

  • New autotuning features include asymmetric autoscaling, in-flight job updates, and intelligent downscaling.
  • Performance boosts with dynamic thread scaling and vertical autoscaling for batch jobs.
  • ARM processor support and SDK performance improvements for better Dataflow operations.
  • ML enhancements with RunInference, model updates without job interruptions, and GPU support expansion. UPDATE

Kubeflow Pipelines v2 enhances ML workflow scalability

  • Kubeflow Pipelines v2 (KFP v2) released, upgrading ML pipeline creation and management.
  • Google's KFP, part of Kubeflow, allows reusable, scalable machine learning workflows.
  • KFP v2 introduces improved Python authoring, Argo-independent definitions, and a better GUI.
  • Features include executable components, nested pipelines, and enhanced ML metadata support. UPDATE

AWS Neuron backs Llama-2 70b model

AWS Neuron SDK now supports the Llama-2 70b model and adds beta support for PyTorch 2.0. UPDATE

Orchestration

Artifact Registry enhances build artifact management

  • Artifact Registry's remote and virtual repository features now generally available.
  • Offers streamlined dependency management across teams with local caching.
  • Enhanced security with Artifact Analysis for scanning vulnerabilities.
  • Simplifies developers' access by consolidating repository URLs. UPDATE

Multi-cluster Gateway for GKE goes GA

  • Google announces general availability of Multi-cluster Gateway controller for GKE.
  • Supports unified load balancing across GKE cluster fleets.
  • Enables blue-green deployments and geo-distributed applications.
  • Enhanced security with integrated advanced features for load balancers. UPDATE

Service Bus updates .NET functions

Azure Functions .NET isolated worker model now supports Service Bus message settlement actions. UPDATE

Provisioning + Runtime

Azure Blob Batch adds Cold Tier

Azure Blob Storage's Cold Tier is now supported in Blob Batch operations, offering cost-effective storage for infrequently accessed data. UPDATE

Azure NetApp Files cools data

Azure NetApp Files standard storage now offers a cool access tier to store infrequently accessed data more cost-effectively. UPDATE

Management Tools

AWS adds 64 new Trusted Advisor checks

AWS Trusted Advisor now integrates with AWS Config for 64 new checks in operational excellence and other categories. UPDATE

NPW Insights this week

  1. O’Reilly technology 2024 landscape
  2. Forrester’s 2024 CIO challenges
  3. LinkedIn Apache Beam case study
Access NPW Insights here