Hyperconverged data centers were once a niche technology that mostly appealed to organizations with specialized needs, such as streamlining the management of small and branch offices. Today, many enterprises are now recognizing the value of transitioning their conventional data centers into hyperconverged facilities. Gartner reports that by 2018 hyperconverged integrated systems will represent as much as 35 percent of total converged infrastructure shipments by revenue, up from a low-single-digit base in 2015.
Hyperconvergence shares some similarities with converged systems, says Charles King, president and principal analyst at Pund-IT. "Both are highly integrated, highly virtualized solutions," he notes. Yet converged systems generally incorporate specific vendors' hardware components, optimized for specific applications and workloads. "In contrast, hyperconverged solutions are clusters that typically leverage commodity components to support software-defined environments and related functions," King observes.
"Hyperconvergence delivers a radical simplification of the IT infrastructure," says Jeff Ready, chief executive officer of Scale Computing, a virtualization and convergence provider. The approach consolidates all required functionality into a single infrastructure stack running on an efficient, elastic pool of processing resources, leading to a data center that is largely software-defined with tightly-integrated computing, storage, networking and virtualization resources. Hyperconvergence stands in contrast to a traditional converged infrastructure, where each of these resources is typically handled by a discrete component that serves only a single purpose. "Hyperconvergence takes the headache out of managing the infrastructure so that IT can focus on running the applications," Ready says.
As more organizations move toward hyperconverged infrastructures, an array of long established and startup technology providers are stepping forward with solutions. Current market leaders include Nutanix, VCE, Scale Computing and Hewlett-Packard Enterprise (HPE). Dell, Lenovo and VMware are also building a presence in the hyperconvergence marketplace.
Each of the market players is approaching hyperconvergence from a slightly different angle. HPE, for example, promises that it can get a hyperconverged system up and running in as little as 15 minutes. Each system, built on industry-standard HPE ProLiant servers, arrives at a data center preconfigured with scalable computing, storage and networking components, along with virtualization software, inside a single 2U chassis.
Nutanix, meanwhile, focuses its hyperconverged infrastructure efforts on systems that are built using web-scale technologies and architectures that originated in large Internet and cloud companies, such as Google, Facebook and Amazon.
VCE says its VSPEX BLUE hyperconverged infrastructure appliance makes virtual initiatives simpler, easier and faster across a broad range of environments and business needs. The company also offers VCE VxRack System 1000 Series, a hyperconverged offering that is designed to complement the company's converged infrastructure systems, providing a hyperconvergence pathway for companies that already have a converged data center.
"VCE, in conjunction with VMware, Nutanix, Pivot3, Sphere3, Simplivity and Scale Computing, is coming out with a wide range of solutions that allow customers to start small and linearly add appliances as needed to meet business requirements," says Courtney Burry, senior director of product marketing for end user computing with VMware. "These types of solutions are extremely popular with customers who are very distributed, as well as for customers who up to this point could not make virtualization — especially desktop virtualization — work due to the complexity of getting it up and running.”
Hyperconverged infrastructures help enterprises that are struggling to cope with increasingly complex IT infrastructures that were bolted together over several years or decades. "Key benefits include faster deployment, lower capital expenditures and operational expenses, and improved IT staff productivity," says Steve Harris, vice president and general manager of Dell Federal. Scale Computing’s Ready agrees. "Hyperconvergence provides a fast and cost-effective way to escape an overcomplicated monstrosity of an infrastructure that has been created over time," he says.
A hyperconverged infrastructure also opens the door to evaluating more powerful, efficient and cost-saving technologies. "If an organization wants to experiment with or deploy software-defined private clouds or a virtual desktop infrastructure, hyperconverged systems provide the ideal means of getting up and running quickly, causing minimum stress to IT staff and resources," King says. "That's especially true for midsize companies that might lack the technical staff or know-how to build a solution from the ground up."
Harris notes that most hyperconverged infrastructure adopters will have to face the challenge of learning an entirely new approach to deploying, operating and maintaining a data center. "They also will need to pay close attention to hardware configuration and sizing to support the specific workloads running in the hyperconverged environment," he says.
VMware’s Burry notes that storage growth and performance tuning create additional pain points for some adopters. "If a cluster runs low on storage but not compute, a customer will still need to upgrade overall compute capability by adding another appliance," she explains. "Likewise, if they want to tune storage disk configuration for a particular application, this can also pose challenges."
Retraining IT staff presents yet another potential, although perhaps relatively minor, adoption obstacle. "Getting and keeping IT staff up to speed on a new solution is absolutely critical," Pund-IT’s King says. "Although factory integration and optimization take much of the complexity out of deploying hyperconverged systems, they're still technologically complex systems that require staff to be trained for optimal use."
Organizations are discovering that a hyperconverged infrastructure, built with a solid technology foundation, can be an excellent addition to traditional servers and standalone storage for a wide range of virtualized applications and workloads that do not require the robustness or scalability of a traditional storage area network architecture. Paul Miller, vice president of converged data center infrastructure marketing for HPE, says that while training is still necessary, hyperconvergence's simplicity and resilience reduces the need for a deeply skilled IT staff. "Customers are looking for something coming into their data centers and remote offices or branch offices that they can power up and start deploying virtual applications almost instantaneously."
Miller notes that that the extensive number of use cases already powered by hyperconvergence is still growing. "The top use cases that we see are small virtualization farms where customers are looking toward general purpose, multiple applications — they want to be able to spin up VMs very quickly," he says.
VDI, collaboration systems, safely isolated test and development environments, and simplified backup and disaster recovery operations are just a few of the top hyperconvergence use cases.
"Hyperconvergence ensures that companies can get up and running much more quickly, that they can linearly and predictably scale out, and that they can streamline management and drive down expenses," Burry says.
Pund-IT’s King, however, warns that the market is growing so rapidly that adopters need to investigate hyperconvergence solutions carefully to make certain that vendors can back up their promises. "The underlying software can dramatically impact the overall price of the solution, so it's important to compare software options," she says.
Harris concurs, noting that companies deploying hyperconverged infrastructures need to be certain that their solution incorporates the most advanced software running on a proven hardware platform with predictable performance for a given workload and environment. "Global service and support, supply chain management and financial services are also important, and will give organizations the ability to implement these solutions quickly and with confidence," he says.
To learn more about how a software-defined data center can deliver agility and efficiency, read the CDW white paper “The Modern Data Center for a Digital World.”