Virtualization is a widely accepted technology and many data centers are at or nearing 100% virtualization. However, many more mid-sized and SMB companies hesitate to grow their virtualized environments because of complexity and expense. Some of these pressures include:
- Limited IT budgets. Companies adopt virtualization to save money by commoditizing servers and simplifying migration and applications. However, virtualization can be expensive with ongoing capital purchases plus licensing and maintenance contracts.
- Minimal, generalist IT staff. Most virtualized environments require a good portion of administrative time. This means dedicating a specialist IT admin, or requiring a generalist IT admin to add the virtualized network to other responsibilities. Losing skilled employees can mean a long recruitment cycle, in the meantime paying contract workers or overworking staff to keep the virtualized environment running.
- Expensive technical support. Generalist IT needs access to training and support, and even specialist IT will need support during upgrades and slowdowns. Even if the virtualization support organization is knowledgeable and willing, their help never comes cheap.
Solving the Problem with Hyperconvergence
IT does not have to live with virtualization complexity if they are willing to adopt hyperconverged systems. At first this suggestion seems counter-intuitive: isn’t hyperconvergence one more expense and point of complexity?
It’s not. It’s new to a lot of environments, and if you have a very simple setup it may be overkill. But if you are committed to server virtualization, then look to hyperconvergence to simplify your computing environment and save money at the same time. Keep in mind too that hyperconvergence is not only for server virtualization but also serves VDI initiatives and improves workload performance.
Hyperconverged systems take a hypervisor, computing, network, and storage components and place them into a single appliance or cluster. Most of these systems are sold as preconfigured appliances but there are also reference architectures so VARs or end-users can build one themselves. Not only does architecture save money on buying all of these components separately, it also saves on management and support time and costs.
- Easy, near-limitless scaling. Hyperconverged systems not only converge computing and storage elements into single stacks, they also integrate multiple stacks. Adding new nodes for performance and capacity is a simple and non-disruptive matter: not something that a storage array can easily do.
- Native hypervisor. Running virtualization through the hyperconverged stack gives a fully integrated and simplified integration with servers, storage, networking, and management. Not all hyperconvergence vendors support all hypervisors, but all of them support VMware and most of them support Hyper-V. There is growing support for KVM and XenServer as well.
- Simplicity. Since hardware and software management complexity are hidden under the covers, even generalist IT successfully manage a fully virtualized network. Employee churn, while never fun, is far less destructive because it’s simple to train new people on the hyperconverged system. It also takes far less time to manage.
- Highly cost-effective. Instead of buying and managing separate servers, network components, storage, hypervisor, and storage management, the customer buys a single highly scalable platform and centralized administration. Cost savings over separate purchases, licensing and maintenance contracts can end in big cost savings.
- Better business continuity. On top of these benefits, in-built replication and integration with backup products make data protection and business continuity a much more efficient and effective procedure.
There are good hyperconvergence solutions at several different price points. Over time even the most expensive of the products will save a lot of money, but the least expensive choice can be the best fit for a smaller environment.