With traditional IT infrastructure, components like servers and storage arrays are purchased in isolation from each other. Each item has a different replacement cycle, meaning that servers may be swapped out for new ones in one year with data centers the next year. This gives companies a lot of freedom but gives IT teams a lot of headaches around compatibility. The solution for many businesses is converged or hyper-converged infrastructure.
Converged and hyper-converged infrastructure vs. traditional infrastructure
When your business relies on traditional IT infrastructure, your technology staffers need to manage and install discrete components within your IT system on an ongoing basis. Integration with vendor-specific hardware interfaces is challenging and can introduce subtle failure points into the infrastructure. In contrast, converged and hyper-converged infrastructure start off with a vision of the IT system your company actually needs.
These methods involve installing and managing compatible, vendor-specified hardware and software. These so-called “hypervisor” platforms recognize, configure, pool and manage all the resources within a company’s IT infrastructure automatically. This provides a unified view of all the available resources and makes their management vastly more simplified and reliable than before. The first key decision to make is whether a converged or hyper-converged infrastructure is better for your organization.
Converged infrastructure explained
Converged infrastructure systems, which are hardware-based, start from a single building block that comprises computing functions, networking, storage and server virtualization. The major benefit of this style is that your IT system rack is fully optimized and your hardware and software work well together with minimal interoperability issues. That’s because all products are normally pretested and preconfigured prior to deployment.
Each of the components in the building block is a discrete component. This means that a component can be used for its intended purpose: The server can be separated and used as a server, just as the storage can be separated and used as functional storage. But they are also set up to be compatible with the rest of your company’s hardware and software.
There are two standard approaches to building converged infrastructure:
- Building-block approach: Fully configured systems – including servers, storage, networking and virtualization software – are installed in a large chassis as a single building block. The infrastructure is expanded by adding additional building blocks.
- Reference-architecture approach: This version allows companies to use their existing hardware, like routers, storage arrays and servers to achieve the same result.
“The simplicity of simply adding a fully configured and tested infrastructure block makes it easier to expand and maintain the network without needing to spend a lot of time reconfiguring the various components,” said Bharat Badrinath, head of product marketing for BI/QuickSight at Amazon Web Services (AWS). “The blocks effectively snap together similar to the colorful Lego-brand building blocks found in a child’s toy box.”
Server virtualization is a way of making a physical server act as multiple virtual servers, like how you can create VLANs from a LAN connection. The benefit of virtual servers is that they can run independently of each other, in effect making one very powerful server appear as multiple servers, with each one having configurable processors, memory and network resources. This reduces costs and the amount of energy needed to run the system.
Converged infrastructure pros
Like all technology solutions, converged infrastructure has its pros and cons. Consider these advantages:
- One control panel: It’s easy to check the current overall performance of your entire network.
- Agility: It’s also easy to load new apps and programs onto converged infrastructure and then grant access to specific users.
- Faster network and cloud connections: Thanks to the ability to dynamically manage available network resources, your digital operations will be speedier.
- Scalability: Integration of new hardware is close to plug and play as long as it’s compatible with the hypervisor technology you’re using.
- Lower support and maintenance costs: Software on the platform recognizes, manages and tunes the components, alerting you to faults before they occur. They’re also able to compensate for any individual component failure by substituting other virtualized hardware in its place.
- Lower upfront costs than traditional IT: Less software and hardware are needed, which reduces initial capital expenditure. Fewer items of equipment also deliver lower electricity usage charges.
- Reliability: Hypervisor software manufacturers publish lists of compatible hardware, meaning that certified hardware has been thoroughly tested for long-term reliability. This gives you the confidence that you have a “battle-tested” component that will integrate seamlessly with your IT system.
Converged infrastructure cons
Weigh these disadvantages when exploring converged infrastructure:
- Predefined configuration: Convergence hardware and software systems have predefined templates and concepts on what constitutes a network, a virtual compute unit or a storage network. You can’t deviate from these standards, so if you have a need that falls outside these requirements, then that’s tough luck.
- Restricted hardware choice: There may be an item of hardware that you really want to add to your system, but you might not be able to if your vendor doesn’t offer it.
- Patch support: You update software patches on the vendor’s timetable rather than yours. Patches must be updated in preconfigured systems in order to maintain support.
Hyper-converged infrastructure explained
Hyper-converged infrastructure is, in comparison, software-based. These setups are often used on private clouds, where all computing, network and storage components are virtualized. The hypervisor creates flexible virtualizations (for example, remote desktops for users) with which you (and the hypervisor itself) can manage computing, networking and storage functions more efficiently.
Because storage is now purely a software service, there is no need for expensive storage area network (SAN) or network-attached storage (NAS) hardware in the hyper-converged infrastructure. You can use less expensive hardware because the hypervisor convergence software takes over their management.
A data center is a company’s shared IT and equipment used for storing and processing data and other assets.
Hyper-converged infrastructure pros
Hyper-converged infrastructure has its own set of upsides that you should take into consideration:
- Cheaper hardware: Systems use commodity hardware and user licenses for virtual machines start at a few dollars a month, depending on volume.
- Easy deployment: Your vendor will install and configure the storage and management software as well as integrate the whole system for you.
- Full integration: Each item of hardware and software and the system itself have been designed to work together and can be configured from the same control panel.
- Flexibility: You can provision and customize private clouds within minutes as all components within the system are virtualized. You don’t need to worry about the installation of servers, patching SANs or servers into the network, or individual configuration of components.
- Fewer staff: When a company deploys hyper-converged infrastructure at all business locations, the need for remote or branch office IT staff diminishes.
Hyper-converged infrastructure cons
The downsides to hyper-converged infrastructure include the following:
- Stuck with one vendor: You’re locked into your provider’s specifications on your infrastructure capabilities even if you wish to use another vendor whose capabilities may be more suitable.
- Difficult to scale incrementally with low cost: It’s easy to scale up, but you’ll be purchasing additional network resources, memory and computer processing power by the node. What if you just need extra memory?
- Lack of control over operating expenses: If your new infrastructure relies heavily on the public cloud on top of any private cloud, you’ll be spending money you shouldn’t be spending on hardware for cloud storage and processing. Some of the software on your system may also be very resource-hungry, further increasing cloud costs. [See Cloud Computing Can Increase Productivity and Profits.]
How to choose between converged and hyper-converged infrastructure solutions
Converged infrastructures are more popular among companies that might need to scale their systems quickly and want to spend less per unit of growth. But many enterprises tend to prefer hyper-converged infrastructure for life cycle and cloud apps, big data analytics, and app development environments.
Smaller businesses that want to move away from traditional IT systems choose hyper-converged infrastructures more than converged ones. Over time, many organizations recognize the additional benefits of hyper-converged setups in terms of scalability and management, especially if they’re planning to grow and extend their IT infrastructure.
The solution that’s right for your company will depend on a number of factors – namely, the extent of your current IT needs, your budget and your expectations for the future as the business grows. Whenever shopping for technology, it’s wise to explore different vendors and compare packages. (For example, see our comparison of the best cloud hosting services and our overview of highly rated cloud storage services.) Then you can make an informed decision on the best option for your business based on your company’s preferences and what the top providers are offering.
Source interviews were conducted for a previous version of this article.