The hybrid cloud balance: knowing when to switch between public and private

Join the leaders in AI and data at Transform 2021 on July 12 for the AI ​​/ ML Automation Technology Summit. Register today.

Over the past few years, industry analysts have discussed the phenomenon of companies considering removing their workloads from the public cloud. In fact, a recent argument that the market capitalization of large-scale public software companies is weighed down by cloud costs and hundreds of billions of dollars has caught the interest of several business executives. It’s easy to misinterpret this as a prediction of an impending public cloud exodus, which I doubt will be the final turn of events. The data shows that only a modest number of companies – a 2019 Gartner survey puts that number at 4% – have actually repatriated (or need to repatriate) their public cloud workloads to a private cloud solution.

My personal view is that the public cloud is essential for digital transformation. It remains one of the most important areas of opportunity for organizations and is virtually the only proven way to scale a business quickly and reliably. And yet, the cloud approach that promises the most value for businesses is hybrid, both public and private, leveraged for the right reasons and at the right time in a company’s lifecycle.

Often, the disillusionment with the public cloud, and therefore the advocacy for repatriation, stems primarily from a poor estimate of the cost savings associated with migrating to the cloud. Contrary to popular belief, the cloud promise of affordable compute and storage is not the most powerful driver of operational efficiency for businesses, especially as growth often slows with scale, as growth often slows down with scale. Unit costs are piling up and the short-term decrease in efficiency begins to sound the alarm bells. Not all workloads are suitable for the public cloud, either. Very often the problem is not the cloud itself, but poor planning and management of the workload as well as misplaced goals. For example, companies that think they can just move and shift their on-premises workload to the public cloud are often disappointed to find that the path to sustainable value isn’t exactly mapped that way.

However, there may come a point in a company’s hybrid cloud journey where, depending on the motivations of the business, it makes sense to scale the business to rely on a public or private cloud. This is when the business is significantly mature in the cloud and may consider optimizing its workload for a number of reasons:

To reduce costs: From an efficiency perspective, it takes full automation of operations built on top of the cloud computing and storage infrastructure layers to leverage the cloud. Simply relying on reductions in initial capital expenditures as well as flexible infrastructure operating costs commensurate with growth in operations places an unsustainable burden on the business as users and scale operations are increasing. This is a tricky situation because the business cannot scale, at first, unless it is in the public cloud, but once it reaches a certain scale, it cannot afford the expense. routine if its IT and business operations are not automated. Another important source of value in the cloud is ecosystems, that is, the relevant platforms and markets that promise companies disproportionate efficiency that cannot be accessed by competitors who may not be running the cloud. same way.

If, however, enterprise scale is the primary motivation for a business to adopt public cloud, when it reaches critical scale, the business may consider repatriation to private cloud, in part. or in full, to control costs. Dropbox offers us an interesting example of a business changing course. Dropbox left the (mostly) public cloud in 2016 to place all of its data in three colocation data centers owned across the United States. In the first year, the company saved US $ 92.5 million on direct billing; even after factoring in the costs of constructing the new facilities, the savings reached nearly US $ 75 million. But to reach 500 petabytes, Dropbox first relied on the public cloud.

The public cloud also offers the possibility of exploiting a whole data economy beyond its own corporate data. Exploration to bring the best of public cloud services to private cloud is now underway on a large scale. This will allow businesses to gain cloud experience and realize the benefits while retaining control over their data to meet governance and data residency regulations.

Beat latency and improve uptime: Public cloud service downtime is rare but can create large-scale disruption when it does occur. At that point, the corporate clients can do nothing but wait. But with a private cloud or on-premises data center, a business has more control over uptime and downtime. For example, healthcare providers that store patient records in the public cloud and data from patient monitoring devices could be retrieved on demand from the private cloud. Latency can also be an issue, especially if the user base is massive and geographically distributed. It is precisely to overcome this that Netflix operates on a hybrid cloud model, hosting its content and user database in a public cloud, while delivering content locally to users through its private content delivery network.

Safety comfort: From a technological standpoint, the public cloud is highly automated for security, which means less human intervention and fewer errors. Cloud security can offer specialized options that are otherwise out of reach, due to cost, for many businesses. Often, cloud-based security services are preconfigured, and if the business prefers the system to be configured differently, there may not be many options. This, on occasion, pushes businesses towards the private cloud, especially when it comes to regulatory environment, reporting requirements, or data sensitivity – think banks and healthcare organizations.

Lack of skills: It takes in-depth skills in provisioning, cloud architecture, and performance reporting, to name a few, to effectively manage workloads when running a private cloud. Those who rely on the public cloud are getting a lot of help on this front due to vendor automation. For example, Google’s Anthos is a managed applications platform that extends the services and engineering practices of Google Cloud to any environment, even outside of the Google Cloud Platform, so businesses can modernize them. applications faster and establish operational consistency between them. So, for companies facing a skills shortage, the private cloud or the on-premises environment may not be viable options.

Obviously, repatriation is not a “mass” option, nor is it the default move to a defined stage of a company’s cloud maturity. At all times, the right workload needs to be hosted, for the right reasons, at the right time, in the right cloud, public or private. The value of automated AI-driven operations, data interchange and software as a service made possible in the public cloud should not be easily discounted. However, every organization will do well to incorporate repatriation into their long-term contingency planning as a possible response to future changes within their own business and also in the public cloud.

Ravi Kumar is president of Infosys.


VentureBeat’s mission is to be a digital public place for technical decision-makers to learn about transformative technology and conduct transactions. Our site provides essential information on data technologies and strategies to guide you in managing your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the topics that interest you
  • our newsletters
  • Closed thought leader content and discounted access to our popular events, such as Transform 2021: Learn more
  • networking features, and more

Become a member

Margie D. Carlisle