Hyperconverged infrastructure (HCI) is a combination of servers and storage into a distributed infrastructure platform with intelligent software to create flexible building blocks that replace legacy infrastructure consisting of separate servers, storage networks, and storage arrays. More specifically, it combines commodity datacenter server hardware with locally attached storage devices (spinning disk or flash) and is powered by a distributed software layer to eliminate common pain points associated with legacy infrastructure.
Hyperconvergence: how it works?
HCI converges the entire datacenter stack, including compute, storage, storage networking, and virtualization. More specifically, it combines commodity datacenter server hardware with locally attached storage devices (spinning disk or flash) and is powered by a distributed software layer to eliminate common pain points associated with legacy infrastructure. Complex and expensive legacy infrastructure is replaced by a distributed platform running on industry-standard commodity servers that enables enterprises to size their workloads precisely and to scale flexibly as needed. Each server, also known as a node, includes x86 processors with SSDs and HDDs. Software running on each node distributes all operating functions across the cluster for superior performance and resilience.
Hardware platform configurations are available to fit any workload by independently scaling the various resources (CPU, RAM, and storage) and can be provisioned with or without GPU for graphics acceleration. All nodes include flash to optimize storage performance, and all-flash nodes are available to deliver maximum I/O throughput with minimum latency for all enterprise applications.
In addition to the distributed storage and compute platform, HCI solutions also include a management pane to enable you to easily administer HCI resources from a single interface. This eliminates the need for separate management solutions for servers, storage, storage networks, and virtualization.
Defining hyperconvergence: the birth of hyperconvergence technology
Datacenter infrastructure has been designed around SAN Storage since the 90’s to protect data and to power critical databases, and became pervasive with the explosion of virtualization in the early 00’s.
But as organizations have grown increasingly dependent on technology, traditional SAN-based infrastructure can no longer keep up with IT needs. It’s complex, unwieldy, and can’t scale as flexibly or efficiently as IT teams need to keep up with changing business priorities.
The world’s largest web companies faced the realities of traditional infrastructure’s limitations long before the broader market, and developed distributed systems technologies to meet their scalability, reliability, and operational efficiency challenges.
In 2009, engineers from several of these web scale companies realized that the technologies they had developed to solve their own operational challenges were applicable to the market at large. The realities of bringing these technologies to enterprise computing required a new approach, and the concept of HCI was born.
Today, HCI is the infrastructure of choice for companies that want to stay competitive and evolve with the changing realities of the technology landscape. While the actual date and person who first coined the term hyperconvergence can be up for debate, Nutanix was the first technology company to bring to market an HCI-specific product in 2011 called Complete Cluster.
Hyperconverged infrastructure FAQs
How does hyperconverged infrastructure help with IT efficiency?
HCI reduces your datacenter footprint by reducing typical infrastructure stacks down to scalable building blocks with compute, storage, and networking built in. And this drastically reduced footprint enables you to run the same infrastructure at the edge as in your core datacenters, resulting in additional efficiency while improving resiliency and performance.
Separate servers, storage networks and storage arrays can be replaced with a single hyperconverged infrastructure solution to create an agile datacenter that easily scales with your business. Hyperconvergence makes administration much easier, enabling you to manage all aspects of your infrastructure from one place, all while reducing complexity by removing compatibility problems between multiple vendors.
If resources become scarce, you simply call your vendor, ask for more servers and software licenses, and deploy them with a few clicks. The infrastructure should be all but invisible to application owners. They shouldn’t have to worry about underlying infrastructure; they should only be focused on their workloads.
What is the difference between converged and hyperconverged infrastructure?
Converged infrastructure (CI) is a different way of purchasing traditional infrastructure and is typically pre-integrated by a vendor or Systems Integrator. Despite pre-integration, CI is built on the same hardware-centric components, and it doesn’t remove organizational silos or solve the problems related to traditional infrastructure.
Now let’s define hyperconvergence. Hyperconverged Infrastructure (HCI) completely re-thinks the way infrastructure can be designed, purchased, deployed, managed, and expanded. HCI is deployed on commodity hardware with all of the intelligence in software and is architected from the ground up to automate the tedious tasks that traditionally plague IT while providing extensive insight and control over the environment. These are fundamentally different architectures that result in drastically different outcomes in terms of business agility, application availability, performance, security, and cost efficiency.
For more information check out Converged vs Hyperconverged Infrastructure.
Is hyperconvered infrastructure technology scalable?
Similar to public cloud services, HCI technology enables IT teams to start with what they need today and scale incrementally to precisely meet application demands. With HCI, you can non-disruptively scale out your environment with modular building blocks as your business needs grow. In contrast, with traditional infrastructure, each tier is sized based on specific needs. In particular, storage is deployed on large monolithic storage arrays that are complex to design and deploy, and often slow down as more applications are added. Once an array fills up, the only way to add more storage is to deploy another large array that has to be managed separately. This dynamic leads IT teams to try to plan for 3-5 years so they can avoid getting into this situation.
Can hyperconverged systems simplify storage?
Data is growing at 50% or more per year, and that data is stored on block, file, and object storage. New requirements for visibility and control are increasing demands on storage administrators. And cloud storage has become an important tier that must be considered in any storage architecture. But traditional storage infrastructure can’t keep up with the demands caused by these new realities. It’s siloed, which creates complexity, limits flexibility, and reduces utilization.
Traditional infrastructure lacks sufficient visibility into the data to support the new compliance and control requirements. It was originally designed in a time before cloud - making adoption of cloud-like capabilities really difficult. HCI breaks down silos and pools all resources into a single resource that’s easy to manage and control. The more “invisible” infrastructure can be the better, and HCI extends that invisibility into the storage domain. With HCI, you can include a variety of nodes in a cluster that make sense for your needs at that point—storage-heavy nodes when you need storage, CPU-heavy nodes when compute is needed, or anything in-between.
What applications can be run on hyperconverged infrastructure?
In the past, HCI started with use cases like VDI and ROBO (remote or branch office). That dynamic has rapidly changed as more and more users of HCI solutions have made their systems available with more and more production and datacenter workloads, even as they prepare their resources for the future.
Some prime examples of what apps run on HCI:
- Databases: Oracle, SAP HANA, Microsoft SQL Server, MySQL, PostgreSQL, IBM DB2, and many others
- Business critical applications: Virtualized server applications like Oracle E-Business Suite, SAP Business Suite, Microsoft Dynamics, Epic, Meditech and more, supported on all major hypervisors
- Big data: Splunk, MongoDB, Elastic, Hadoop and more
- Cloud native: Build and deploy cloud native applications with Kubernetes, Docker, Puppet, Chef
- Virtual desktop infrastructure (VDI): Citrix, VMware Horizon
- Remote office and branch office (ROBO): Print and file servers, office services, custom applications
What kind of hardware is used with hyperconverged infrastructure environments?
The components of an HCI system include a distributed infrastructure plane and a distributed management plane.
The distributed infrastructure plane runs across a cluster of nodes delivering storage, virtualization, and networking services for guest applications - whether they’re VMs or container-based apps. The management plane lets you easily administer your global HCI resources from one place and one view. It eliminates the need for separate management solutions for servers, storage networks, storage, and virtualization. HCI solutions are 100% software-defined - zero dependency on proprietary hardware. HCI provides the choice of a wide range of appliance and server platforms from multiple server vendors.
Related articles:
Building a Business Case for HCI
Top 20 HCI Questions Answered
