Are you looking for hyperconverged infrastructure?
Maybe. I suspect that few of you are actually sending out a request for proposal (RFP) for hyperconverged infrastructure. That’s not how, in my experience, most of you who are IT professionals think about solutions. What you do think about is the fact that you have a budget that you have to live within, you have requirements to support growth, change, and modernization that you have to meet, you have initiatives for management efficiency, green IT, performance, availability, and recoverability of applications – but you don’t have a requirement for hyperconverged infrastructure. Hyperconverged systems are an approach that may satisfy many of your requirements, but they are not the requirement and there’s no budget line item for hyperconverged systems.
That is not the same as saying that IT professionals aren’t buying hyperconverged infrastructure or looking for the benefits that this approach delivers. You are – it’s a hot topic right now.
Analysts have tried to provide structure and guidance for how you should evaluate converged solutions. Gartner analyst, George Weiss, developed a taxonomy that includes six categories of converged systems. On one side are Integrated Stack Solutions that come pre-packaged with compute, networking, storage, virtualization, and the application. On the other side are build-your-own-systems (BYOS) leveraging software-only hyperconverged Integrated Systems (HCIS) approaches. Some would argue that the Integrated Stack Solutions provide what may be the most efficient approach for delivering a single application. At the same time, the BYOS approach promises maximum flexibility in terms of application, hypervisor, operating system, server, networking, and storage options. Ultimately, for you, the IT decision maker, the approach you adopt will depend on how you budget and allocate costs, your approach to support and vendor management, your concerns regarding supplier lock-in, your readiness to invest in retraining staff, and the extent to which you believe you can anticipate and control infrastructure requirements and adapt to technology shifts going forward.
Many of StorMagic’s larger customers have an architecture team with an in-house lab where they can test multiple permutations of servers, virtual storage appliance software, disk drives, memory, and hypervisors. These customers tend to run multiple applications and know they may add additional applications in the future, but at the outset, they deploy identical systems to hundreds or thousands of sites as part of a branch-office modernization initiative. Because they have a good handle on their workloads and because they have the in-house resources to evaluate and optimize their infrastructure specifically to their workload, they tend to adopt the BYOS approach to hyper-convergence. They can buy the precise server, memory, and storage configuration that meets their specific workloads, without over-provisioning. It is often, for these customers important to squeeze every ounce of waste out of the purchase, while leaving open the option for upgrades and changes going forward.
Other StorMagic customers want a solution for a single application. The customer may not have even considered that it’s StorMagic inside, but what they want is the high availability that StorMagic SvSAN delivers at an affordable cost. What the customer ultimately is buying in this scenario is what Gartner describes as an Integrated Stack Solution. These solutions are delivered by the builder of a physical application appliance such as a medical PACs system or an industry-specific analytic application.
In the past couple of years there have been no shortage of announcements regarding new hyperconverged infrastructure (HCI) solutions. Partnerships have dominated the news, including Cisco partnering with StorMagic. The fruits of these partnerships ranged from BYOS to Integrated Stack Solutions, with vendors offering tailored HCI solutions towards SME data centers, ROBO, enterprise data centers and everywhere in-between. These solutions are designed to offer you, the buyer, confidence that solutions have been tested and will be supported as a joint solution with, typically, a single point of contact and a single lead on support. This approach is ideal for organizations that have limited in-house testing resources, a desire for a single support contact and an-appliance-like approach without the vendor lock-in. Typically, while they are delivered and supported by the systems vendor or their distributor, if the systems vendor falls out of favor, the customer can switch systems vendors without severely disrupting how the systems are managed.
Even if you are relatively new to the IT profession, you have probably had the opportunity to witness at least one or two supplier relationships that have been affected by vendor consolidation. Relationships tend to have a long tail, even when they are ending. So, customers should not have an immediate cause for alarm. But it is a good time to evaluate your approach going forward. You may decide that a software-defined storage approach, that enables the use of any server hardware, any hypervisor, and any storage infrastructure as part of the hyperconverged offering, provides greater flexibility and less supplier risk.
If you’d like to try SvSAN, you can download a free 10 day trial and get started today. If you’re deciding whether to go for a BYO or appliance-based HCI deployment, why not check out our dedicated white paper?