A worthwhile reminder of the importance of the true meaning of hybrid – when one often hears in conversation phrases such as ‘it’s a bit of a hybrid solution’ when someone is referring to a DIY solution that isn’t perhaps as elegant or well-engineered as it could be, but it gets the job done. Let’s be in no doubt, hybrid is a good thing – very much so in today’s digital world. Firstly, we have the hybrid workplace. Already a thing in its own right since the dawn of our connected world, the seemingly distant days of Covid and lockdown accelerated the idea that, providing they could access their applications and the outside world, folks could work from anywhere, at any time. Working from home five days a week remotely may not be for everyone, or allowed by many employers, but there are certain advantages for both the company and the individual. Similarly, 9-5 Monday to Friday in the office has its upsides and downsides for everyone. Hence, many organisations now understand the importance of the hybrid workplace – allowing their staff, within certain well-defined boundaries, to vary their place and hours of working. Workplace flexibility and agility – sounds like a plan! Secondly, we have hybrid IT – the idea that IT can be supplied to a business and the people who work for it in a variety of different ways. On-premises hardware and software solutions; the company’s hardware and software assets relocated to a colocation provider’s data centre – they can use their expertise to ensure that these valuable company assets are well looked after and always available; or everything can be removed to the cloud – outsourcing taken to its logical conclusion. As with the workplace example above, most organisations now understand the pluses and minuses of each approach, and have almost certainly arrived at some kind of a hybrid IT solution – placing their resources where it makes most business sense to do so. Thirdly, we have the hybrid cloud. No matter how many of a company’s IT resources are ‘in the cloud’, there’s every chance that a one size fits all approach is something of a compromise. Different applications may well best be served by different clouds. Then there’s the idea that, if you are using a colocation provider to host some of your IT assets, it would make a great deal of sense to also access the cloud services which are located in the same data centre location. And talk of data centres brings us on to fourthly – the hybrid data centre. Yes, some companies still have all their IT assets housed in one central data centre, or across a variety of computer rooms and server cupboards on their own premises; but plenty are also using colocation data centres to host more and more of their IT assets – as already mentioned – these companies are experts in the power, cooling and overall facilities management and maintenance required to ensure that your IT applications are available 24x7x365, or as they like to put it five 9s availability (99.9995) – although I suspect we are already in the realms of six or seven 9s; and, as already discussed, cloud data centres have the necessary IT infrastructure to provide applications to organisations. So, the likelihood is that most companies have a hybrid data centre policy. Sticking with hybrid data centres for fifthly and sixthly, one could argue that the use of large, centralised data centres, alongside smaller, regional facilities and, increasingly (and predicted to explode at some stage) very small, edge data centres is another definition of a hybrid data centre architecture. The idea being that the right applications and data are placed in the optimum location to provide optimum performance levels for the user – whether that be employees, partners or customers. Sixthly, and this might be stretching a point – but there’s also a growing trend to consider the geographical location of a data centre. There’s something of an overlap when it comes to the fifth point, but, in terms of sustainability in particular, there is increasing consideration being given to where the data centre is located in terms of reliable and sustainable access to power, water and cooling solutions, for example. We’ll leave it there for now. Before we leave the data centre, we can’t ignore some other potential hybrids (seventhly and eighthly, I think). First up, the idea that either within the same data centre or across two or more data centres, different workloads are best served. With the advent of AI, this point is growing in importance. A single data centre will need (radically) different infrastructure to host equally well what we might call traditional work applications and, for example, the Large Language Models (LLMs) which are the foundations of the AI world to come. More likely, AI applications will be placed in AI factory data centres – built to provide AI-optimised infrastructure. This split between traditional and AI applications may well require different cooling solutions. Hence, more and more talk of hybrid cooling. Air-based solutions continue to dominate – with free cooling the ultimate objective, but liquid cooling is gaining traction as the only way to cope with the heat generated by AI chips inside servers. And here we encounter hybrid number nine – different approaches to liquid cooling. So, we have not just hybrid cooling, but potentially also hybrid liquid cooling – maybe a mixture of direct to chip and immersion cooling. So, cooling brought specifically to the hot chips, alongside larger assets, such as servers, being carefully immersed/surrounded by liquid. We could also extend the idea of hybrid to the three core IT hardware categories – storage, networks and compute. After all, storage within an enterprise tends to be broken down into primary, secondary and archive layers, provided by SSDs, hard disks and tape (yes, there’s still a role for this technology!). And networks can be split between fixed and wireless solutions, as well as copper and fibre cables. As for compute, well the range of devices used to access applications ranges from small, handheld devices, through tablets, laptops and desktops, to the huge, high performance supercomputers that drive the digital future in so many industries. If we allow storage, networks and compute to be given hybrid status, this should very much help us to understand and, more importantly, embrace the less familiar hybrid digital world to which organisations aspire. After all, I imagine few, if any of you, consider yourselves to be users of hybrid storage or hybrid networks. You simply use the right storage for the right job. Similarly with networks (and compute). Ultimately, hybrid is about taking the way in which you (comfortably) provision your IT hardware and extending it across every single aspect of your digital infrastructure estate, to provide the agility, flexibility, speed and scalability required to optimise the user experience. A final question for you: if the dictionary definition involves only two types of plant or animal, and many of the examples I’ve given above involve at least three ‘types’ of solution to consider, do we need to start using the term ‘multi-hybrid’ to best describe what’s happening right now?! Rightly, you will conclude that the name doesn’t matter, so long as you understand why it matters and how to get there. Next article: A data centre can be as sustainable as you want A data centre can be as sustainable as you want >