Billing options


Choose hourly billing and pay for what you use. Starting at USD 0.04 per hour.

→ See all pricing


Choose monthly billing for consistent pricing and budgeting. Starting at USD 28.75 per month.

→  See all pricing

Types of virtual server instances (VSIs)


Choose between monthly or hourly balanced, balanced local, compute, memory or GPU options:

  • Balanced — ideal for common cloud workloads requiring a balance of performance and scale
  • Balanced local — ideal for large database workloads requiring high I/O performance and low-latency performance; 100 GB primary standard on all local storage configurations
  • Compute — ideal for CPU-intensive workloads
  • Memory — ideal for memory-intensive workloads
  • GPU — ideal for accelerated computing workloads, such as machine learning, deep learning and high-performance computing (HPC)

Dedicated servers

  •  Single-tenant offering with rapid provisioning for further control and flexibility in VSI deployments
  • Choose up to 56 cores and 484 GB RAM
  • Monthly or hourly options


Whether you refer to it as spot pricing or transient servers, it's all about running your flexible workloads on unused capacity at a fraction of the cost. When the capacity needs to be reclaimed, the transient servers will be deprovisioned on a first-on, first-off basis.

Transient servers are ideal for the development and testing of applications and other nonproduction workloads that don’t require constant uptime — allowing you to save without stretching your budget.

Reserved servers

Reserve a set of up to 20 specific virtual machines in advance for guaranteed capacity in the data center of your choice. This option is ideal for sustained workloads and enterprise-level capacity challenges. Choose a one-year or three-year term with consistent monthly payments and reclaim your reserves at any time.

Dedicated hosts

Single-tenant dedicated hosts help with control over workload placement by ensuring your workloads are placed appropriately based on your requirements. You can also view resource consumption — core, RAM and local storage — for maximum control over workload management. Dedicated hosts can also support network throughputs of up to 20 Gbps. Learn more about configuring for higher network performance.


What are profiles?

When you provision IBM Cloud® Virtual Servers, you can select from supported families of profiles. A profile is a combination of vCPU and RAM that can be instantiated quickly to start a virtual server instance. In the IBM Cloud console, you can choose from popular profile configurations or select from a list of profiles that best fit your needs.


Best for common cloud workloads that require a balance of performance and scalability, the balanced profiles with network-attached storage provide higher performance because resources aren’t oversubscribed.


Best for moderate to high web traffic workloads, compute profiles are best for workloads with intensive CPU demands, including production batch processing and front-end web servers.


Best for memory caching and real-time analytics workloads, memory profiles are best for memory-intensive workloads, such as large caching workloads, intensive database applications or in-memory analytics workloads.

Get started on IBM Cloud Virtual Servers in minutes

Can virtual servers match your workloads? Provision a public, dedicated, transient or reserved IBM Cloud Virtual Server and explore the options.