By this point, probably millions of words on the Internet have been dedicated to articulating the benefits and drawbacks of the public cloud. As the title of this post suggests, I’m not here to convince you one way or the other: The benefits of the public cloud are real (as you probably know), and if the public cloud is not somewhere on your roadmap, it ought to be.
At the same time, there are very real challenges and potential drawbacks, and risk mitigation must also be a key part of your cloud strategy if you are to have any hope of success on your cloud journey.
There. Now that that’s out of the way, let’s talk about how to cloud. I’m going to talk about what is going on with cloud in telecommunications, share some trends, and offer a couple of suggestions. My hope is that maybe this can be instructive for how business and IT leaders, both in the telecom space and outside of it, should think about the public cloud and architecting their analytic workloads.
One of the bigger stories in telecommunications is that operators have been extremely slow to embrace the public cloud and partner with the hyperscale cloud providers. There are a number of reasons for this, including the typical security and performance concerns that accompany multi-tenant environments.
But a big reason was that migration has been slow, largely because “lift-and-shift” – the practice of picking a workload up, dropping it into a cloud environment, and hoping it worked – often didn’t succeed. During the initial rush to the public cloud, and even today, lift-and-shift was the quickest path out of the data center; but it was never optimal, and it was usually not the desired end state for those legacy apps.
Even if the application was written to function in a completely virtualized environment, and the underlying hypervisor didn’t matter all that much, those workloads were not architected to take full advantage of the cloud. They were monolithic and self-contained. Rewriting or refactoring those applications for the public cloud, or replacing them with Software-as-a-Service, was more effective, but that was also time-consuming and expensive, required new skills, and the effort carried its own risks.
Yet the cloud is critical to a lot of what telecom operators are trying to accomplish, particularly in the era of 5G. Consider these advantages in the context of telco transformation:
- Object storage – 90% of new data coming into operators will be unstructured or semi-structured data from the Internet of Things (IoT), and that data will need a place to land. The growth of data lakes will be a key trend in the coming years.
- An ecosystem of tools – More important than simply collecting and storing data, the cloud represents an ecosystem of tools that can help telcos leverage those massive volumes of data.
- Elasticity – Rapidly spinning resources up and down to more closely align with customers’ consumption is a critical component of managing the cost of delivering a high quality service as efficiently as possible.
The telco migration to the public cloud was and is an inevitability. However, like most organizations, they will almost assuredly be multi-cloud and hybrid – a mix of on-premises and cloud environments, with valuable data stored across the organization.
There is also evidence that they’ve learned from their slow crawl towards the public cloud. There is great interest in containers among telcos, and a major driver of that is interoperability. While the public cloud makes a lot of sense for many workloads today, their dominance is not guaranteed, and in the hypercompetitive race to dominate the 5G landscape, telcos can’t afford to take years to adapt to the next technological shift. Interoperability accomplishes three important things: 1) removes the risk within their infrastructure decisions today, 2) enables their multi-cloud and hybrid cloud reality, and 3) mitigates their risk in the future.
This logic is not limited to infrastructure – applications should feature the same interoperability. Analytics platforms should enable multi-cloud and hybrid cloud. That’s because for many large organizations, ingesting large volumes of data, using a central repository, and adopting an all-cloud or only-cloud architecture is simply not feasible.
At the same time, our business realities demand analytic insights based on our entire data footprint.
So that’s the big takeaway. You are going to the cloud, and the outcomes are worth it. Just make sure that the cloud, and all of the technologies that make up the ecosystem, truly serve you and the interests of your business, today and far into the future.
A great way to realize the power of a cloud-optimized architecture without leaving your data center is with Vertica in Eon Mode, deployed on-premises. We, along with our growing list of object storage partners, are here to help!