Datacenter teams are finding that the levels of scale, simplicity, and agility needed to support complex digital transformation initiatives simply cannot be achieved using the long-standing practice of buying individually managed silos of datacenter resources, or just following the “check box” next to things like application programming interfaces.

 

As a result, IT organizations around the world are shifting their resources away from standalone servers, networking, and storage systems in favor of converged infrastructure solutions that can be centrally managed with tools that offer new levels of automation.

 

The adoption of converged systems has grown quickly, with annual spending now more than $13 billion worldwide. It is expanding faster than the larger datacenter infrastructure market, which is a clear indication that converged systems are driving real benefits within the datacenter.

 

An important element of this growth is the relatively recent emergence of hyperconverged infrastructure (HCI), which IDC considers a subset of the $13-billion converged systems market and the next phase of the market’s life cycle. HCI solutions deliver the proven benefits of traditional converged systems, but do so through a software-defined, scale-out architecture.

 

HCI solutions are built as clusters of commodity servers (x86) that provide an abstracted pool of capacity, memory, and CPU cores that are used as the foundation for server-centric workloads (the hypervisor, virtual machines, and applications), as well as storage-centric workloads (e.g., data persistence, data access and data management).

 

Although not yet widely realized, the hyperconverged market is entering a new phase of maturity. HCI solutions are now increasingly targeting mission-critical workloads, are larger in scale, and are used by a wide range of companies. Based on past IDC surveys, business applications (e.g., ERM, CRM, supply chain management, financial management, payroll/accounting) were the most common workloads running on HCI solutions. Other common applications running on HCI include collaborative and content applications, structured data analytics, and structured data management applications. IDC expects the use of HCI for mission-critical business applications to expand considerably.

Hyperconverged Solution Requirements Going Forward

While the types of workloads running on hyperconverged solutions are a good indication of how far the HCI market has quickly progressed, more must be offered to drive further market expansion. Modern HCI solutions must close important feature/capability gaps that exist with traditional datacenter infrastructure, enable an IT department’s shift towards public and private cloud infrastructures, and support an increasing need for organizational transformation within the datacenter.

The following are the most important attributes and capabilities a modern HCI solution must provide to ensure relevance as organizations push for expanded HCI use cases within their datacenters:

  • Application Performance Protection. With expanded use of hyperconverged infrastructure comes the need for HCI solutions that support an increased density of primary workloads. Thus, one of the most critical feature gaps that HCI solutions must address is support for predictable and guaranteed performance levels within such environments. With increased workload density comes the increased risk of resource contention. For hyperconverged solutions to thrive during this next phase of hyperconverged adoption, they must consistently deliver sub-millisecond response times and support hundreds of thousands to millions of IOPS. This is creating the need for a workload protection engine to address “noisy neighbor” issues and ensure that performance can be consistently delivered as required, no matter what else is going on in the system.
  • Integrate with Multiple Public Clouds. Modern HCI solutions must provide a private cloud platform that can be easily integrated with hyper-scale public cloud providers to create a seamless hybrid, multi-cloud experience. Features required to achieve this include:
    • Use of service catalogs that support on-premises resource and workload deployment
    • Support for highly portable workloads and automated bursting
    • Ability to seamlessly move data between on-premises, private clouds, and public clouds to support the ever-changing needs of a diverse workload portfolio
    • Seamlessly collapse management complexity to better support the lifting and shifting of applications from private clouds to trusted hyper-scalers and back again as business requires
    • Support for containers to accelerate development of new services
    • Support for chargeback of resource utilization
  • Provide a Common Data Fabric for Private & Public Clouds. IDC believes hyperconverged solutions have become an ideal platform for on-premises, private cloud deployments thanks to their scale-out, software-defined, highly automated architecture. But it’s important to recognize that the role of customers’ cloud-based IT solutions is to ingest, deliver, and exploit data no matter where that data is created or lives. Thus, a truly optimized hyperconverged-based private cloud should be considered a component of a broader, hybrid public/private cloud ecosystem that provides a “lingua franca” (or data fabric) that supports common data services for efficient placement, movement, and use of data across a true hybrid multi-cloud environment. Such a data fabric allows IT departments to incorporate newly deployed HCI solutions in their datacenters without creating additional management silos.
  • Enable Organizational Transformation. IDC notes that modern hyperconverged solutions are also increasingly looked upon to support organizational transformation within the datacenter. Therefore, modern HCI solutions must support consolidation of datacenter jobs/roles so that one administrator can take on the responsibility for virtualization, compute and storage. This consolidation of roles will help to free up time within the IT team for more innovative projects and ultimately help shift human capital to other critical parts of the datacenter that drive innovative, new revenue streams, and customer touchpoints rather than just maintaining status quo.
  • Scale Resources Independently. Many HCI solutions scale by requiring an additional node that would increase compute and storage resources together. This is commonly referred to as linear scaling. The problem with this is no datacenter grows in a linear way across all the pools. There are times when there is need for additional compute or additional capacity, but not both. To address these increasingly diverse needs, HCI solutions must evolve into a more elastic solution that grows with business requirements, not simply with IT vendor configuration requirements. If an environment becomes “storage intensive” or “compute intensive,” the infrastructure must be able to support this without stranding resources or requiring users to add a full node with more compute or capacity than is required.

To learn more about how hyperconverged infrastructure is meeting the stringent demands of mission-critical and hybrid cloud environments, including NetApp’s HCI portfolio of solutions, download the IDC white paper, The Evolution of Hyperconvergence and NetApp’s Role in this Rapidly Expanding Market, sponsored by NetApp.

Eric Sheppard

Eric Sheppard is a research vice president of Server and Storage Infrastructure at IDC, where he covers enterprise storage systems, enterprise storage software, converged systems, and hyper converged infrastructure. Eric frequently contributes to primary research and custom storage projects and regularly presents market trends at industry events.