IaaS 101: Types Of Compute

An introduction to IaaS Compute at IBM Cloud

Compute is a complex and essential component of our Cloud platform. In simple terms, Compute refers to the process where computers take data and instructions and execute them in order.

While seemingly simple, Compute plays a crucial role in our modern world and has helped us do everything from landing on the moon to streaming our favorite shows on Netflix. Our platform and others, such as AWS, Microsoft Azure, Oracle, Alibaba Cloud, Digital Ocean, and Google Cloud, provide customers with various computing services tailored to their needs and budgets. These platforms allow customers to offload the significant costs of purchasing, operating, securing, and maintaining the hardware required to run their applications and businesses.

Data centers play a crucial role in making these platforms operational. Located in large buildings around the world, they house numerous rows of servers connected through extensive networks of cables, switches, and routers. These facilities ensure consistent access to cooling, power, and internet connectivity, all managed by skilled administrators who ensure everything operates smoothly. Customers can rent as much or as little as they need and focus on building their businesses, not maintaining infrastructure.

As more companies transition to the cloud, the most common customer use cases include:

1. Building Cloud-Native Applications: Applications specifically designed to run consistently across private, public, and hybrid cloud environments or extend legacy applications into the cloud.

2. Expanding Capacity: Companies leverage cloud infrastructure to overcome limitations posed by on-premises hardware availability and to take advantage of rapid scaling capabilities.

3. Easier Access to Cloud Services: Businesses can more easily access and utilize cloud services, products, and features without relying on complicated workarounds.

4. Modernizing Legacy Monolithic Applications: Organizations are updating old applications to benefit from the latest infrastructure and containerization technologies.

5. Enhancing Resiliency: This includes implementing additional data security and disaster recovery measures.

6. Cost Efficiency: Finally, cloud computing often proves to be a more affordable option compared to maintaining on-premises infrastructure.

Compute at IBM Cloud is broken up into different specialized teams. Our specific IaaS (Infrastructure as a Service) compute design team works on VPC Virtual Servers, VPC Bare Metal Servers, Images, Platform Services, Autoscale, and Dedicated Host, and we even support workload teams such as HPC and SAP when needed. The other teams that round out the compute catalog are Power VS, IBM Z (Hyper Protect Virtual Servers), Kubernetes (IKS), Red Hat OpenShift (ROKS), and finally, Code Engine.

At the heart of our computing products are two foundational technologies: Virtual Private Cloud (VPC) infrastructure and our first-generation Classic infrastructure. VPC infrastructure (including VPC Virtual Servers and VPC Bare Metal Servers) is designed for modern, multi-zone cloud-native applications. It is particularly suited for regulated and highly secure industries due to its enhanced security and privacy measures. VPC infrastructure also supports modernized hybrid and multi-cloud environments. Classic infrastructure is often used to migrate legacy applications to the cloud, extend existing on-premises applications, and perform high-performance cloud computing tasks such as gaming, finance, and database management.

Bare Metal for Classic

Our first-generation infrastructure is an entire physical server dedicated to a single customer. These devices are highly secure and stable but tend to be the most expensive option. Additionally, due to their physical nature and the availability of components in various locations, setting up and taking down these servers requires more time.

Virtual Servers for Classic

Virtual servers for Classic are created by dividing and virtualizing a classic bare metal server using a hypervisor. Allowing customers to rent portions of a physical machine or the entire machine itself. This setup is ideal for volatile workloads (due to its quicker provisioning speeds) and for low-intensity computing applications.

Virtual Servers for VPC

Virtual servers for VPC are created by dividing a bare metal server and virtualizing the instances using a hypervisor. These servers operating on the latest hardware within an isolated VPC network. Compared to traditional classic servers, deploying a virtual server in a VPC offers several advantages, including faster provisioning times, improved performance, greater control over the environment, enhanced security and privacy, and reduced interference from other users.

Bare Metal for VPC

Bare Metal Servers for VPC has transformed our popular product, Bare Metal Servers for Classic, to better meet customers’ needs in highly regulated industries and evolves the offering to align with a cloud-native application deployment model. Designed for mission-critical workloads in the financial services and healthcare sectors, Bare Metal Servers for VPC offers dedicated servers that deliver high performance within a scalable network, allowing users to easily extend or migrate their on-premises workloads to our cloud with minimal retooling and granular controls, in just minutes.

Analogy — If we were to create a neighborhood, these offerings would serve as the foundation for everything we build.

Once customers select a foundational technology, they can consider two additional platforms if needed. The first is VMware, which is an older and more familiar technology. The second is containers, a newer method that operates on virtual machines and offers a modern approach for running cloud-native applications, such as Kubernetes and OpenShift.

VMware

VMware, which pioneered virtualization is an external company that has ongoing partnerships with many cloud providers. It assists customers in migrating their existing on-premises VMware environments to the cloud of their choice, allowing them to access modern services. On IBM Cloud, VMware runs on Bare Metal for Classic servers and utilizes a specialized hypervisor called ESXi.

Analogy — If we look at the following products as real estate built upon our foundation, VMware offers customers a house or a condo with various customizable options, depending on their needs. VMware tends to be more expensive than container-based solutions and is generally a viable choice primarily for customers with existing VMware runtimes.

Kubernetes

IBM Cloud Kubernetes Service (IKS) enables users to run managed Kubernetes clusters on various types of underlying hardware. Kubernetes is an open-source container orchestration platform that simplifies the deployment and management of multiple containers. It automates tasks such as deploying, scaling, healing, and managing containerized applications, making it easier to handle them effectively.

Analogy — IKS would be considered a motel to start (but if users are willing to invest time and materials, it can be developed into a luxury hotel). It is a highly cost-effective and no-frills approach to running containerized applications with a robust ecosystem of open-source tools you can add to it.

OpenShift

Red Hat OpenShift on IBM Cloud (ROKS) is a Kubernetes-based container platform that enables users to run managed OpenShift clusters, allowing developers to create, deploy, and manage applications across hybrid and multi-cloud environments.

Analogy —OpenShift can be likened to a 5-star hotel. It enhances user experience on top of Kubernetes, allowing developers to leverage Kubernetes products fully.

Bonus: Dedicated & Shared
Many of our products have an option to choose between a dedicated environment and shared environment. A dedicated environment is a complete server or host that gives the customer full control over its resources and usage. These environments are high-performance and highly secure but come at a premium cost compared to shared environments.

A shared environment allows customers to rent a portion of a server or host shared with other users. This option is very cost-effective and suitable for many customers, depending on the applications they need to run in the cloud.

Analogy — Dedicated environments would be considered a house: they are isolated from their neighbors (not sharing walls), and the owner has complete control over what they can do with the property. On the other hand, shared environments resemble condos, hotels, or apartments. In these cases, tenants share resources and are close to one another, often making them much more affordable than a house.

Another essential part of our compute catalog includes the Power and Hyper-protect products, which cater to specific use cases such as artificial intelligence, applications involving highly sensitive data (confidential compute), high-performance computing, and extending on-premises Power workloads.

POWER systems Virtual Servers

Power Systems is a family of servers from IBM that utilize the Power processor architecture. These servers are designed to handle the most data-intensive workloads, which exceed the capabilities of x86 architecture. While customers initially used Power servers primarily on-premises, IBM Cloud now offers this enterprise-grade platform for hybrid and multi-cloud deployments.

Hyper-protect Virtual Servers

These are s390x servers that provide a confidential environment to protect your data. They are designed to address the topmost security concerns, and not even IBM Cloud administrators have access once the resource has been provisioned.

Wrapping this up
While this space can seem daunting, vast, and complex, remember you are not alone. Our team collaborates to understand, create, and enhance our customer experiences. We hope this article has provided a helpful introduction to Compute at IBM Cloud.

– Written by Austin Edwards, Maria Gillcoat, and Josef Bodine
– Illustrations by Austin Edwards
– Article updated in January 2025

To learn more about IaaS Storage and IaaS Networking on IBM Cloud, you can read our other articles, IaaS 101: What is — Enterprise Storage and IaaS 101: What is — Networking