The ROI of HyperConvergence
These days we hear so much talk about hyperconvergence, and mostly we talk about reliability and spend very little time thinking about the cost of acquisition (until we go to buy it, of course) and the return on that investment over time. Outside of improved reliability, what should we be looking for as the overall business use case for hyperconverged approaches?
Of course, hyperconvergence is a broad concept and not limited to a singular product, vendor, style or approach so we must remain somewhat aloof and consider that these concepts are very general, potentially do not apply to a specific product but these tend to be the advantages of the approach and why hyperconvergence, beyond the strict architectural advantages, tends to be beneficial over time.
The key to hyperconvergence value is in collapsing complexity and disparity into repeating bricks that are sometimes identical physically but are at least highly similar logically. This logical similarity is what makes for the biggest factors, outside of reliability, in making hyperconvergence valuable for us from a business perspective. By reducing complexity to the IT team and business we have a huge opportunity to reduce the biggest impact on any businesses IT: human costs.
IT is always hampered by the high cost of IT staff, generally reflected in an inability for IT to get enough done as there is always so much more to do than IT has the resources to tackle. Hyperconvergence approaches this problem by taking the typical storage and compute layers of core IT infrastructure and collapsing them into a single entity and further by merging disparate systems (this is an organizational decision of course, but it is the expected cultural shift) into a single platform. This simplification of the logical architecture can be significant in changing how IT works and how focus can shift from platform support as a core function to enabling IT to focus primarily on areas in which IT represents an opportunity for competitive advantage.
Hyperconvergence represents one of the more significant commoditization movements within IT today. Much as how cloud computing commoditized much of infrastructure in the enterprise space, hyperconvergence is doing this for the SMB space.
In most organizations, platforms and systems are the two largest support centers and require the biggest expenditures both in time and often in purchasing as well. This is where expertise is generally focused and where gaps are most prevalent. The amount of knowledge and experience needed for this arena is extreme and often nearly unobtainable in a realistic way for non-enterprise IT environments which then causes internal IT to either be forced to tackle things for which they are not prepared, to over staff internal IT or to rely on consultants for those functions. The alternative, in the platform space, is to simplify platforms to reduce or even essentially remove these needs.
In many cases we can justify hyperconvergence up front through a reduction in TCO (total cost of ownership.) Hyperconvergence alternatives, such as the Inverted Pyramid design with dedicated storage and compute layers, are often more expensive to obtain initially and implement even before we consider their administration, maintenance and risks costs down the road. In most cases, hyperconvergence is actually the less expensive approach out of the gate or is, at the very least, very price (of acquisition) competitive.
But the big value of hyperconvergence is an operational one. By removing some of the biggest, most complicated layers from our platform stack we reduce support costs for the indefinite future. And in most cases we reduce the complexities of higher layer management (network and compute) significantly as well making our entire infrastructure easily manageable by IT staff that might have struggled to manage even a single layer of an alternative approach's design.
Further support is clarified as hyperconvergence encourages the merging of vendors down to a single one which reduces IT coordination costs (primarily in terms of time) so that there is a single support channel to call (one throat to choke) when something goes wrong, expansion is needed or consultation is desired. Support is simplified and time is not wasted getting the wrong vendors involved, finger pointing between vendors or finding the right support resources. This generally speeds problem resolution, while almost always speeding engagement.
Of course, hyperconvergence brings us flexibility and simplified growth patterns that allow for delayed investment in more cases than other system designs and a more long term strategy rather than investing into an expected rip and replace generational strategy like we have been stuck using in the past.
Through simplification and commoditization of the architecure, improved reliability of design and support improvements, hyperconvergence can not only compete on TCO with traditional designs but then wins dramatically over time as system costs and IT value cost are reduced, competitiveness increases and ROI shows a significant win.