Announcing the Public Preview of Azure Compute Fleet

This post has been republished via RSS; it originally appeared at: Azure Compute Blog articles.

Today we are announcing the public preview of Azure Compute Fleet, a new service that allows you to manage and deploy thousands of virtual machines (across a mix of SKU’s, VM type, Availability Zones, and pricing models) with a single API call.


Embracing the "Fire & Forget-it" model, we are introducing Compute Fleet to automate your deployment, management, and monitoring of virtual machines (VMs) without requiring intricate code frameworks. Tell us what you need, capacity and instance wise for Standard and Spot VM and Compute Fleet will create both Standard and Spot VMs from a customized SKU list, tailored to your workload requirements with the capability to maintain the Spot VM target capacity in the fleet.

Compute Fleet allows you to accelerate your access to Azure’s compute capacity in a region by launching a combination of Standard and Spot VMs at the lowest price and highest availability. You can attain superior price-performance ratios by utilizing a blend of diverse pricing models, including Reserved Instances, Savings Plan, Spot, and Pay-As-You-Go (PYG) options.

Deploying Compute Fleet with a pre-defined fleet allocation strategy, allows you to optimize your mix of VMs for capacity and cost efficiency or a balance of both. This will alleviate concerns about complexity associated with determining optimal VM pricing, available capacity, managing Spot evictions, and SKU availability.

There will be many ways you can use this product, whether by running a stateless web-service, large batch jobs, a big data cluster or Continuous Integration pipeline. Workloads such as financial risk analysis, log processing or image rendering can benefit from the ability to run thousands of concurrent cores.


Key Benefits

  • Deploy up to 10,000 VMs
  • Mix multiple VM types (Standard /Spot)
  • Mix multiple Azure pricing models
  • Mix multiple VM SKUs/sizes
  • Expedite capacity access
  • Hyperscale using Spot VMs
  • Maintain Spot VM target capacity
  • Optimize for price, capacity or both

Customers of Azure Compute Fleet:

  1. Large enterprises needing to deploy and oversee thousands of VMs concurrently while maintaining an optimal balance between price, capacity, or both.
  2. Organizations, especially those rooted in cloud-native practices, requiring flexibility in deploying and managing latency-sensitive workloads across various SKUs, zones, Spot instances, and other Azure pricing models.
  3. Customers aiming to amalgamate pricing models and VM types within a singular application to expedite outcomes or scale up to tenfold more efficiently on Spot instances within the same budget  

Available Now:

You can use Compute Fleet today in the following Azure Regions: East US, East US2, West US, West US2

We will be available in more Azure regions soon!

Learn more about Azure Compute Fleet

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.