When should you deploy a software-defined data centre?
The top techies already know the buzz around SDDC. They are also aware that, unlike other shifts in enterprise IT, SDDCs constitute an evolutionary change in data centre design and management.
As such they could prove a risky undertaking if their potential impacts are not understood thoroughly, However, as Tikiri Wanduragala notes, some vendors have already anticipated the need to by using platforms that can upgrade existing infrastructure and become the building blocks of SDDC environments.
For many organisations, this is still a work in progress. To answer the C-suite’s expectations, IT decision-makers may have to produce reasons that spell out why the SDDC route is not (yet) suitable for every organisation.
For instance, in addition to significant deployment work, an on-premises or co-located SDDC implementation calls for new disciplines to be brought into the IT team. This is because functions like workload allocation of software-defined compute, storage and network resources require IT teams to program, or set policies, for the logic of that allocation, rather than using manual processes supported by system management tools.
And the advent of ‘self-service’ infrastructure means that IT managers are tasked with enabling these needs directly. This requires them (and their teams) to predesign offerings for consumption by application developers – a capability lacking in many IT departments, according to Gartner.
Business drives deployment
The short answer to the “When should we?” question is governed as much by business factors as technology strategy. The SDDC proposition offers ROI opportunities, only achievable through closer collaboration between IT leaders and business owners. This typically brings cultural upheaval for an enterprise, but it’s upheaval that’s been on the cards for some time.
Considering the scale of disruptive change SDDCs could cause, Gartner advises that organisations be modest regarding their initial forays into SDDC adoption. Enterprises are more likely to achieve results if they have smaller, ‘greenfield’ projects to which SDDC models can be deployed. Examples might be business initiatives that are limited in scale and scope, and come with clearly specified requirements that allow for more precise SDDC product alignment.
Another key aspect of the evolution toward SDDC is that the transition affects not just the server and storage layers, but all aspects of the data centre infrastructure. As Lenovo’s William Scull points out, control of network hardware is , thus enabling networks to be created and managed in ways similar to those already used to manage virtual machines and virtual storage.
As network management is migrated to centralised software control, it makes the fully software-defined data centre an operational reality. But an organisation that’s just made an investment in connectivity hardware may want to sweat those assets for a while before looking at software-defined networking.
Think before you switch
As this IDC white paper outlines, data centre owners have little alternative to change. For enterprise facilities, it’s a case of upscale or watch the business falter; for service provider facilities, it’s a case of evolve or see customers switch to rivals with software-defined techniques built into their infrastructures from the outset. Both sectors have found conventional data centre technology designed for yesterday’s needs rather than tomorrow’s requirements.
It’s in this vital respect that Lenovo’s difference wins out. To support the customer journey towards SDDC, Lenovo introduced two new brands – ThinkSystem and ThinkAgile – to deliver innovation where organisations most want it. Technology providers like Lenovo and its partners are already helping early SDDC adopters, and they have the experience and insight that can be shared with clients beginning their SDDC journeys.