At its recent annual Build conference, Microsoft introduced the public preview of Azure Compute Fleet, a new service designed to streamline the provisioning and management of Azure compute capacity. This service optimizes performance, cost, and scalability across various virtual machine (VM) types, availability zones, and pricing models. Compute Fleet allows users to efficiently access Azure’s compute capacity in a specific region by launching a mix of Standard and Spot VMs at the best price and highest availability, using different pricing models like Reserved Instances, Savings Plan, Spot, and Pay-As-You-Go (PYG).
See also: Microsoft Azure AI and ML Offerings
Optimizing VM Deployments with Compute Fleet
Azure Compute Fleet allows users to deploy up to 10,000 VMs with a single API call, integrating both Spot and Standard VM types. This flexibility helps users achieve better price-performance ratios by maintaining the Spot VM target capacity within their fleets. Using a predefined fleet allocation strategy, users can optimize their VM mix for capacity, cost efficiency, or a balance of both. This approach addresses concerns about determining optimal VM pricing, managing Spot evictions, and ensuring SKU availability.
See also: Microsoft Azure: A Quick View
Comparisons with AWS and Google Cloud Services
Azure Compute Fleet joins similar AWS and Google Cloud Platform (GCP) services for managing large VM deployments. AWS provides AWS Systems Manager Fleet Manager and Amazon EC2 Fleet, which allow centralized server management and collection of EC2 instances across multiple environments. GCP’s Google Cloud Deployment Manager automates creating and managing Google Cloud resources.
Currently, Azure Compute Fleet is available in the West US, West US2, East US, and East US2 regions, with detailed documentation available for users. This new service from Microsoft aims to enhance the efficiency and cost-effectiveness of deploying and managing large-scale compute resources, making it a competitive addition to the cloud services landscape.
Elizabeth Wallace is a Nashville-based freelance writer with a soft spot for data science and AI and a background in linguistics. She spent 13 years teaching language in higher ed and now helps startups and other organizations explain – clearly – what it is they do.
