COVID-19 and the push to hybrid cloud
In the last few weeks, all of our lives have been changed by the coronavirus pandemic. The lucky among us are healthy, and have stable jobs with organizations that will weather the storm. All of us have family and friends that are not so lucky.
How will this impact the world of work? Well clearly a huge takeup in remote access, collaboration platforms, and a need to boost cybersecurity for workers unexpectedly using corporate systems from personal devices.
Lower down the stack, Infrastructure will also be impacted – in teknowlogy Group’s view, receiving a significant boost.
At its most basic, bread-and-butter data centre work has become more difficult. The 2m rule makes two-person jobs difficult. And since colo facilities in some of Europe’s worst-affected markets have already become ‘intelligent hands only”, colo users in UK must expect the same may yet happen over here. At the same time, DC-operators are taking precautions including keeping many days supply of food and water on-site, in case of unexpected and more stringent lock-down conditions.
Yet while getting things done with physical infrastructure is challenged by the coronavirus crisis, the future for abstracted services looks much more promising. Organizations that already run some of their applications on cloud or virtualised platforms know how they perform, and what they are suited to. Many are now considering how to make more effective use of both their virtualised internal resources, and also their external cloud services.
And we’ve been here before, back in 2008 when there was more talk of hosting than of cloud. At the time hosting providers suffered like everyone else as their customers’ businesses failed, or cash dried up for payments. But the customers that survived put more of their IT into hosting data centres, since in a time of crisis, renting IT feels safer than speculative capital purchases.
Today it’s the same only more so. De-risking spend is even more pressing in current times, and with cloud services (unlike hosting) the services can be scaled down in times of reduced demand – and reduced usage results in reduced fees. This is a double-edged sword – some Finance teams will be worried at the prospect of open-ended spend, but most will be reassured that costs can be scaled back at any time, if and when that becomes unavoidable.
Another very attractive feature of cloud and virtualised services is that they are designed with remote management in mind. Besides the obvious health and safety benefit (nobody needs to drive to the data center), there is also a service reliability benefit: these services also enable automation, which typically improves performance, service availability, and once implemented may way reduce costs.
Of course the middle of a crisis is not the best time to be choosing a new platform. For that reason, many organizations are concluding that in the short term, the best clouds to expand into during this pandemic are the cloud services they are already using. Infrastructure teams that had planned a rather measured and carefully staged move to cloud have been reassessing the timescales. For less critical applications, many are aggressively shrinking their timescales, limited only by the number of resources they can afford to assign to migration tasks.
More complex and more critical applications are not so straightforward. Typically they need some degree of adaptation / rewriting to make them cloud-ready, and this process is not easy to accelerate. In any case most organizations are naturally more cautious with their critical systems, although many organizations are looking to do more with virtualisation, which gives some of the benefits of cloud, at much less risk. Again in current circumstances, the right virtualisation environment is probably the one you already use – whether that’s VMware, HyperV, Red Hat or something else.
Finally, many organizations want to tie their infrastructures together in a flexible but remotely manageable way. This is beginning to become possible, using software defined network solutions. At their simplest, these provide SaaS environments (such as Equinix’s ECX) in which IT users can connect IT infrastructures with network and cloud services.
In summary then, the pandemic is constraining organizations’ abilities to physically work on their IT hardware, and this is pushing them to speed up their moves to more modern, software-powered alternatives. This is a change that most were planning anyway – it’s just that this pandemic has given urgency to the need for newer, more flexible platforms.