Purpose-built for private cloud

From day one, our goal was to create a solution that allowed organizations to build on-premises infrastructure that behaved like a true cloud, not legacy data center infrastructure. We built a storage layer first (just as Amazon did with S3), where we perfected the fundamental scaling issues, taking a systems and holistic view of the problem.

After we had delivered our private cloud storage nodes, we added compute nodes and interconnects. The result is HyperCloud: a unique technology that allows you to scale your cloud atomically and horizontally.

Starting from scratch

Building a true private cloud requires completely rethinking how to design and build infrastructure.

First, you cannot build a true cloud with a legacy data center infrastructure. Building a true private cloud is only achievable by using infrastructure designed to be used for the cloud.

This is something the hyperscalers realized more than 15 years ago, abandoning legacy infrastructure at that time, moving to designing, building, and using purpose-built cloud infrastructure ever since.

As we examined the challenge, we ultimately found that there are four core principles that one must commit to in order to build cloud infrastructure.

Natively ephemeral

A true cloud infrastructure is able to independently spin up additional resources—whether that be compute or storage capacity, without effort. Equally, those resources should “disappear” again when not required, able to be consumed by other tenants or services.

To do this, HyperCloud nodes don’t hold their state, instead getting their assignment from the cluster on boot. They pull their configuration and begin working. When a workload spins down, the node’s resources are released and placed back in the pool, awaiting its next service.

This ephemerality means HyperCloud nodes are auto-configured, maintained and hot-swappable.

Atomic peering

Cloud is built to be easily consumed, not built or configured. If you build your cloud from a rat’s nest of generic servers and subsystems, you can never quickly, simply, and effortlessly “consume” it. Eventually, its complexity and fragility will mean it will fail, or you’ll have to stop everything and reconfigure manually.

That is not cloud.

The foundation of SoftIron’s HyperCloud is the atomic cloud peer (ACP). “Atomic” means these units are the basic, immutable building blocks of the cloud. “Peer” means they are all peers—they all contain the critical cloud elements to operate as a unified fleet—true whether the nodes are optimized for compute, storage, interconnect or any other task-specific function.

They contain all the important hardware platform design elements; the “cloud DNA” that enables SoftIron infrastructure to behave as a true private cloud.

Fleet intelligence

A true cloud infrastructure distributes its management across the installed infrastructure, ensuring there is never any supervisor, single point of failure or choke point.

As new nodes appear, they are assimilated into the cloud and automatically configured by the existing cluster. As a fleet sharing common hardware and software management capabilities, the cluster can detect problems, self-maintain and self-heal. When an upgrade is required, a single update is applied to the cluster that updates all hardware and software autonomously and without service interruption.

Robust scale-out

Like a swarm of bees in nature, fleet intelligence and resilience grow seamlessly as the swarm grows, incrementally adding greater redundancy and management resources every time additional nodes are added to the cloud.

This ability to scale out is independent of any one type for compute or storage technology, allowing any and all to be added as the cloud grows, with no theoretical maximum.

Adding nodes is a simple operation, requiring nothing more than racking and stacking the additional nodes.

How do I build a cloud using HyperCloud?

It’s deceptively simple. For smaller installations you’ll only need around half a dozen bits of information to configure your cloud, and you could be up and running that same day.

Work with us or a SoftIron + Co channel partner to define the workloads you initially want to run and the performance and resilience they’ll need.

We’ll guide you in selecting the right configuration of HyperCloud nodes. All the software to deploy, scale, manage and consume your cloud comes with the nodes, so no need to worry about additional complex licensing models or multiple vendors.

Place your order, and we’ll get your build underway in one of our SoftIron advanced manufacturing sites. Your products will be built in either Northern California or Australia depending on your location.

Receive your nodes at your data center or a location of your choosing and we’ll guide you through the install. You can do this alone, with us or a SoftIron + Co partner, or a combination—the more the merrier!

Discover HyperCloud

Take a deep dive into the technology layers underpinning HyperCloud’s reductive architecture.
Deploy cloud at your data centre or co-lo in half a day and (less than) half a rack, with only generalist IT skills.
Integrated management to enable self service and multi-tenant deployments across private and hybrid infrastructure.
Out of the box cloud, with out of the box pricing.
Back to Top