2015 was Cassini Review’s first year, so clearly for us it wasn’t just another year. We started Cassini Reviews to address what we saw as the imbalance in discussion and understanding between the consumption of data centre services; cloud, storage, servers etc. and the production side of data centres. Perhaps it has been our wide-eyed innocence, if you can call grizzled 30 year IT veterans innocent, but it feels as though there have been some pretty significant changes going on under the surface this year.
The year that Governments got it.
Earlier in the year I was talking to Emma Fryer, Associate Director, Climate Change Programmes, at techUK, who has a particular interest in data centres. She felt that Government Ministers were now more aware, not only of the green issues but more broadly about the importance of data centres for the economy. Also, when a Docklands data centre is designated as the second highest terrorist risk in the country you begin to realise the critical infrastructure nature of data centres.
Across Europe we have seen national and local governments offering tax and other incentives to lure data centre operators to set up. There may have been other considerations but there is no doubt that Ireland, Sweden and Netherlands have benefited from such interventions. We haven’t quite seen the venal and very public “give me a tax break or I won’t come” debate that has been going on in Michigan in the past few weeks, but I have no doubt there have been some very hard-nosed negotiations going on.
Being a ‘Bit-Barn’ is no longer an option.
Almost one of the first things we were aware of as we started analysing the market was the blurring of lines between retail and wholesale co-location. At first this manifested itself as a general decline in the average size of large co-location deals. However it was clear that customers wanted a broader set of services and capabilities around connectivity and support. The acquisition of Telx by Digital Realty is the most dramatic indicator that existing models needed to change. It wasn’t about market size…Digital Realty was already the biggest data centre operator in the world…it was about providing a more retail, global inter-connectivity focused proposition to counter the progress of Equinix.
For smaller regional and local co-location players the answer has been to become more horizontally integrated, either by buying IT Consultancies and local integrators or building closer sales and marketing ties at the expense of the traditional neutrality they adhered to.
Again others have built very specific propositions around a particular location like Verne Global in Iceland, Green Mountain in Norway, Hydro66 in Sweden or Volta in the heart of London. Whatever, just providing power and space is no longer a viable option.
Everyone is talking about the Edge
No, U2 are not getting into data centres as far as I am aware! The ever increasing bandwidth requirements of streaming technologies, mobile connectivity and the promise, or is it threat, of the Internet of Things has forced some serious consideration about where we both process and store the transactions and data for these applications.
While in the short term this may provide more opportunities for local and regional data centres the emergence of companies like EdgeConneX and a real re-imagining of what constitutes a data centre could have a profound impact on how the industry looks in years to come. I don’t think we are going to see a move away from the mega-scale data centres, and we are long way from the vision of the individual as a data centre I heard espoused at DCD Converged in London this year, but there will no doubt be opportunities at the Edge.
Key challenges ahead
I’m sure everyone is seeing a challenge in their own area, but as we have talked to people across the industry this year 3 key issues emerge that could slow the progress of the sector in the coming years.
Servers and storage are OK, but networking is the problem child. Whether it is the time and cost involved in laying down new connectivity, or the challenges in bringing new technology like photonic switching to bear, connectivity problems could yet undermine the rosy growth figures many see for the data centre industry.
Access to capital is going to be critical as companies attempt to build out data centre infrastructure. The purchase of InfinitySDC’s Slough data centre by Virtus is a vivid example of what happens if you are or aren’t well funded. Getting the right investors who understand the long term nature of data centre investment and aren’t looking for the quick and large returns that might have been available 10 years ago will be critical to future development. The question is, are there enough of them out there?
Finally there is the skills issue. With a recent survey indicating that the average age of data centre engineers is 54 there is a pressing need to attract a younger generation into our data centres, after all their socially connected, mobile lives depend on them. This also calls for a more considered approach to training and professional accreditation. The industry can’t rely on the education system to provide the flow of talent they need, and the government has enough on its hands without another sector moaning they need to do more. The responsibility lies with each and every data centre company to attract, train and retain the best. Poaching staff from others is not a long-term fix.
Bring on 2016
While I had spent a lot of my career involved in “data centre” business, none of it had actually focused on the underpinning infrastructure. Indeed as the world moved away from mainframes data centres seemed less necessary somehow. Being shown how wrong that view was, and how critical data centres are to the interconnected world we live in has been a genuinely exciting journey for me in 2015. Notwithstanding the challenges above the future looks bright and I can’t wait to get stuck in in 2016.