Is the cloud really ready to serve media?

It’s a perennial question that comes up in various industry panels and presentations and one which will undoubtedly feature heavily at IBC this year.

aThe answer, of course, is that it depends on what you want to use it for. Some areas are widely acknowledged to be very ready (and indeed already supporting complex production environments). Others, typically those requiring more bespoke and base-band video processes, are less viable. Well, for now anyway…

But the topic of readiness isn’t just about technology; it also has to be about the commercials. After all, one of the central purported benefits of the cloud (and a reason for adopting it) is its supposed financial flexibility and cost effectiveness. But does this necessarily hold up?

The first question has to be, ‘compared to what?’ The general comparison here tends to be with traditional approaches where the client has to make some form of upfront commitment, either in terms of capex or volume to achieve certain discount levels. On the face of it, this is where the cloud does appear to be a commercial step forward with many services in public clouds being offered on a pure pay as you go basis. That said, even the AWS and Azures of this world will still provide preferential pricing to those clients prepared to make volume commitments. Furthermore, it’s commonplace for service providers to create the impression of utility via some form of financing where capex would otherwise be required. Typically though, even here the client will carry some form of ‘cost of finance’, however it’s positioned. So whether or not this is true utility is open to debate. First blood to the cloud.

Sadly, however, we can’t call the fight on such broad strokes. While true utility or pay as you go is commonplace among the large public cloud providers, it’s not so commonplace among the traditional media application vendors who are entering the ‘as a service’ space. Now that’s not to say there aren’t very encouraging signs. In fact we’re working with a wide selection of cloud application vendors (e.g. encoders, transcoders and so on) that have very interesting offers. That said, there is still a large swathe of ‘traditional vendors’ where the ‘as a service’ model presents a challenge. It’s open to a (long) debate as to whether or not cloud ultimately means that vendors will make more or less margin; but what is often the case is that they’ll make smaller chunks of margin over a longer timeframe. I think we’d all prefer to get as much of our cash as early as possible in any transaction, so this does present a dilemma for application providers – one which in many cases, you’ll only do when you need to. So on closer assessment, no clear victor here (for now).

But let’s assume everything’s moving in a cloud like direction, which most would agree it is. Let’s also assume you’ve got access to a shiny, dynamically scalable, full utility based cloud solution for your particular need. The question that’s still not answered is, ‘is it cheaper than just buying the product upfront?’ On this point, I don’t think anyone can provide a firm ‘yes’ or ‘no’. If you’re simply dipping into something for a one-off task now and again, the cloud may well be cheaper. If you’re consuming a utility based service for a period of years, then at some point, your total cost of ownership (TCO) will become greater as a result of on-going opex compared to the initial capex. It’s just a case of when… What’s also important here is to consider how you’re using your opex service, as public cloud price points vary considerably depending on things like volumes and the speed with which you wish to complete a process. Generally speaking, unfettered use of ‘as a service’ will typically reduce the timeline over which your TCO benefit exists for opex over capex – and in some cases, extremely quickly.

All in all, it’s hard to give a definitive answer as to whether the cloud is a) commercially ready or b) necessarily better value. It really does depend on the use case and which model you’re comparing it to. If we are to talk in general terms, however, for me, the primary question isn’t about whether the cloud is ‘cheaper’. It’s more about your financial objective. The reason to pursue opex isn’t just about a better TCO. It could be about your access to capital to make an investment (which you may or may not have). It could also be about a broader organisational goal to off-set things like asset depreciation, managing your overall debt and smoothing your cash flow profile. Ultimately, all of these things should be taken into the mix when considering whether cloud is the right route for your business. And if it is a path you choose to take, make sure you have a robust plan in place around how to control it!

Kris Hardiman, Head of Product Management.