What is composable infrastructure?

If you’ve read any industry blogs or articles lately, chances are you’ve come across the term. So what exactly is "composability?"  Why do we need it? Is this the same thing as "converged infrastructure?" I get these kinds of questions often, and I think they’re important.

There are four different categories of infrastructure: traditional, converged, hyperconverged and composable. Since they form a historical sequence, let’s start at the beginning and work our way up to the present.

1. Traditional infrastructure

Definition: Traditional infrastructure is defined as servers, storage and networking switches. These come in all sorts of shapes and sizes, and you can combine them to optimize for a particular workload. 

Benefits: Your data center was, at least at one time, built on traditional infrastructure, as that’s where the industry started. It’s very flexible in running different types of applications.

Limitations: The challenge is that traditional infrastructures tend to be very cumbersome to deploy and manage. Because compute, storage and network run on different platforms, they create many physical islands of highly underutilized resources. The management tools don’t cross those divides, so they also create silos of management, which make it extremely difficult for server, storage and networking admins to work efficiently.

IT leaders complained that this was just too complex. So the industry set out to do things better. Converged infrastructure and hyperconverged Infrastructure both have started on the journey of making IT infrastructure simpler to buy, easier to operate and faster to consume.

2. Converged Infrastructure

Definition: Converged infrastructures bring together compute, storage and networking into a single solution for a particular workload or solution area, such as virtualization or database.

Benefits: This simplifies the purchase and can potentially make the infrastructure faster to consume for the application it’s designed for. Convergence also can reduce the data center’s power and space footprint.

Limitations: While converged infrastructure does have some benefits, it achieves them at the expense of creating a management island surrounding it--in this case, though, it’s an island that’s created around workloads rather than hardware. Furthermore, the management of the servers, storage and networking is often still done largely in silos, even when the equipment physically integrates them.

3. Hyperconverged Infrastructure

Definition: Hyperconverged systems bring together compute, storage and networking in a single solution.

Benefits: Can be very easy-to-consume infrastructure capacity.

Limitations: Hyperconverged infrastructure supports virtual workloads that do not require connectivity to SANs. Physical and SAN-attached applications require a different infrastructure. As a result, hyperconvergence creates a management silo around these systems.

While both converged and hyperconverged approaches have merit, they fall short of the ultimate goal: a single platform with a single operational model for all workloads. In order to make this a reality, the platform must have hardware that can support a broad range of physical and virtual workloads and be configurable through a software-defined approach to match the needs of a given application or workload. This is the rationale for composable infrastructure.

4. Composable Infrastructure

Definition: Composable infrastructure brings together compute, storage and network fabric into one platform, similar to a converged or hyperconverged infrastructure. It also integrates a software-defined intelligence and a unified API to “compose” these fluid resource pools.

Benefit: Rather than being pre-configured for a single workload like a converged or hyperconverged infrastructure, composable infrastructure is customer reconfigurable, through a software-defined intelligence, to become whatever is needed. In this sense, composable infrastructure is actually the opposite of converged infrastructure. As the “converged” part of the name implies, converged infrastructure is the result of configuration that’s done in the past--before it arrives at a customer site.  In contrast, composable infrastructure enables businesses to easily create the infrastructure they need when they need it--enabling the fast reflexes that they need to thrive in a highly competitive environment.

Composable infrastructure is the next evolution in data center computing because it enables IT organizations to create the infrastructure they need with incredible velocity, and it does so while reducing the operational cost and complexity of traditional silo approaches. It’s ideal for traditional workloads and environments, as well as newer cloud and mobile apps. Think of it as your infrastructure as code.

We have lots of great references to help you learn more about composable infrastructure. Here’s a useful overview: Composable Infrastructure – Bridging Traditional IT with the Idea Economy. Or learn about HPE Synergy, HPE’s approach to composable infrastructure, here.

Gary Thome is VP & Chief Engineer, HPE Converged Datacenter Infrastructure.