This article was originally published on Technology.Info.
As part of our continuing strategy for growth, ITProPortal has joined forces with Technology.Info to help us bring you the very best coverage we possibly can.
In the IT business, we often think of convergence in terms of technology, whether it be printer, copier, scanner or voice and IP. We see these trends taking shape in products and in the shifts that are taking place as vendors merge and consolidate to meet the converged reality - or disappear altogether as new suppliers emerge.
Those of us on the vendor side of the industry tend to think much less about the impact convergence has on the people working with the affected technologies, so what skills do they need to add to their arsenal to stay relevant in the new world?
For a long time, IT engineers and operators have been accustomed to working in their area of expertise, usually clustered around silos of technology that have evolved over time. Organisations created silos, and because silos existed, that is where people have spent their time. If a server team wanted storage, it needed to ask the storage team. When it needed network connectivity, it asked the networking gurus. As we move to a world where the software can ask, on the fly, for more of what it needs, the barriers between these silos collapse and all of us must adapt our skills to work around the new structure. In other words, technology convergence forces career convergence.
We see this convergence first hand when looking at how the relationship between voice and networking teams has evolved in response to IP becoming the dominant means of voice and video communications. Every medium to large organisation has had to make tough decisions about how the voice and networking and even server teams would get collapsed into a single entity with responsibility for delivering IP communications or VoIP to end users.
Today, many are experiencing a similar shift as software defined data centres and cloud architectures are designed and deployed. The software defined data centre promises flexibility, end-to-end flow through provisioning and management, as well as the awareness to deploy applications on demand at acceptable service levels and the ability to access - also on demand - appropriate resources from the data centre. Gone are the formal requests to the silo next to you or below you – instead, the application and the orchestration engine is in charge and the IT organisation must build capabilities within each layer of the IT stack to enable this new world.
Software defined data centres require organisations to take a holistic view of their IT structure and restructure their operations team. As a result, there is much more of a need for a true DevOps mindset and skill set. The experts stuck working in silos will be flanked by colleagues that build IT platforms so they can deliver the agility promised by software defined data centres.
The future of storage is also based on a software defined model, so storage teams too will need to be part of the new approach. Even the largest proprietary vendors whose business models are based on customer lock-in now agree that the underlying hardware will be decoupled via storage virtualisation to provide flexibility and dynamism, which will be extended into storage management. The data plane will be separated from the data management layers, and everything will be delivered as software that can define the attributes of the storage in the server, flash, and JBOD on the fly and in response, again, not to formal requests and annual plans but in response to API driven instructions.
Software defined storage (SDS) will extend from on-disk formats through APIs and business models, be widely available and able to work with all major protocols, and be able to serve block, file and object protocols.
To deliver on SDS, people working in storage will quickly find themselves having to broaden their awareness beyond the management and maintenance of their own silo, as the focus shifts to a service that delivers data wherever and whenever it is required in a truly agile fashion and via the best platform for each particular task. We already see this happening in companies from smaller shops (where generalisation may already be the rule) to several 200PB+ deployed storage users. I personally was shocked (and truly impressed) when the SVP of operations at one of the world’s largest telcos recently let me know that he and his senior management had recognised this trend and had already moved down the path of building cross-functional, cross-silo operational teams to assist in building - and then operating - its increasingly software defined data centre.
What this telco is experiencing now, all medium to large IT shops will be experiencing very shortly. Cross-training, reorganisation, team building, shared instrumentation and end user focused SLAs are some of what is required to ensure the storage team, server team, networking team and even the security team on both the engineering and operations sides work together to deliver the promise of software defined data centres.
The shift can be daunting. At one global telco, I was tangentially involved with assessing members of the team for DevOps roles in its huge cloud services push. Out of the 3,500 operators and engineers reviewed, just 32 made the cut. On the other hand, my experience in the convergence of voice and IP taught me (and I think Cisco’s field leadership) that the legacy expertise of the voice teams was invaluable. But it was only invaluable as part of a broader, SLA focused, converged team.
It is not going to be an overnight process. There is probably more uncertainty now over the fundamental structure of IT than there has been for many years, and SDS is merely adding to the potential anxiety. Many storage teams feel like they are under threat, and with good reason. They are; or rather, the old way of working is under threat. But they should draw comfort from the fact they are an integral part of the SDS future and will play a central role on the way to delivering it.
With SDS, the industry is embarking on an evolutionary leap that takes it beyond the boundaries of convergence of individual technology elements, to a future where IT is truly holistic. The onus will be on those teams focused on areas like storage, server, networking and applications to develop a layer of holistic intelligence on top of their existing skills that will provide them with an understanding of how SDS and its elements can be delivered. The end result will be a better IT. By converging your capabilities, you will participate in an effort to truly fix IT so it delivers on competitive advantage and absolutely, positively matters.
Evan Powell is CSO at leading enterprise-class storage and SDS specialist company Nexenta Systems.