Is hybrid cloud right for you? 6 considerations

The engineering axiom "Let the best solution win" doesn't seem to hold true anymore. With the emergence of a new category of competitors, market dynamics, and emerging use cases, the "good-enough solution" may be the winner, especially when viewed as part of a broader macro environment instead of just considering its technical merits.

Case in point: Telco service providers once put together the best solution for voice calls that was fast, secure, and met customers' needs—at the time. By the early 2000s, telco companies were riding high on their fixed-line and mobile networks. They had a great business model that allowed them to offer enhanced capabilities on a predictable cadence (for instance, 2G to 3G to LTE to 4G) to their ever-growing customer base.

Image

The telco service providers also incurred huge capital operational expenditures just to maintain the ability to offer their services. They could offset these costs with revenue from customers using these services. They also charged a premium for additional services such as texting, long-distance charges for interstate or international calling, and TV on mobile devices.

[ For more on the history of mobile architecture, read The road to cloud RAN from 1G to 5G. ]

However, over time, customers switched to third-party applications to make international calls, communicate, and watch video. People still paid service providers but used them solely as a dumb pipe. The service providers wanted to offer these new services to their customers but were constrained by their architecture, operations, and culture limitations.

The extensive networking equipment at service providers was well-engineered and great at what it did (primarily connect voice calls). It also met customers' needs with a guaranteed service level agreement (SLA) of about five 9s (99.999%) uptime.

The fixed-line model carried over to mobile services. The telcos now also had huge capital expenses as they bid for spectrum, ongoing refreshes to their equipment to support new technology, and operating expenses to manage the mobile network. It's been estimated that the 30 largest telcos have incurred more than $900 billion in operating expenditures (OpEx) every year since 2016, while capital expenditures (CapEx) were more than $270 billion in 2020.

Who stole my cheese?

Over the last decade, telco companies saw their profit margins and growth prospects under severe pressure. As consumers, we've become increasingly frequent users of many calling, messaging, entertainment, social media, and other services. The companies offering these services (as well as many companies making the hardware on which these services are consumed) often have many times the revenue, margins, growth prospects, and stock market valuation than the telco companies. These valuations have decreased significantly since their peak in late 2021, but the contrast is still stark.

Telco's role has been largely reduced to being a provider of network connectivity, while the value-added services are increasingly delivered by other companies—few of which were around two decades ago.

Upstarts disrupting traditional businesses are not unique to telcos. It is playing out across different industries, including manufacturing, entertainment, hospitality, and retail.

[ Learn why open source and 5G are a perfect partnership. ]

Getting back in the game

However, telco companies have begun to co-opt many of the principles and practices that made their non-telco competitors successful. Instead of building their own datacenters, they're increasingly using on-demand and pay-as-they-consume infrastructure services to deliver across diverse use cases with minimal CapEx and OpEx. Additionally, by adopting open source software, telcos are gaining the flexibility to mix and match on-premises infrastructure with public cloud infrastructure, all monitored and managed through a single pane of glass with an abstracted API layer.

More specifically, telcos needed to functionally split their software stack in addition to decoupling hardware and software. The telcos started with virtual machine (VM)-based solutions to decouple hardware and software and gain greater flexibility in workload placement. But as demands evolved, it became evident that VM-based approaches couldn't match the automated deployment and scaling capabilities offered by cloud-native technologies such as containers. The cloud-native infrastructure provided the flexibility and scalability to deploy network functions, scale efficiently, and provide isolation for services like network slicing.

Although the above example applies specifically to telcos, other industries can similarly adopt a flexible cloud-native approach that extends from on-premises datacenters to public clouds to the network's edge.

6 benefits of switching to hybrid cloud

The overall benefits of a hybrid cloud approach are not about adopting a hybrid cloud for its own sake but the potential business benefits it can bring. If the business benefits aren't there for whatever reason, then you may not be a good fit for a hybrid cloud-based solution.

What are the potential business benefits? Here are six that should probably lead you to explore the move to hybrid cloud:

• Cloud bursting
• Edge computing
• Data privacy
• Data protection
• Business continuity
• Drive Innovation

[ Learn how to build a flexible foundation for your organization. Download An architect's guide to multicloud infrastructure. ]

1. Cloud bursting

Cloud bursting is a way to provide dynamic on-demand scaling in industries with fluctuating demand for services, such as retail stores during the holiday shopping season or telcos with high service demand at an event like a big football game. Instead of over-provisioning capacity to handle peak workloads or letting the service degrade, cloud bursting enables companies to offload computing to public cloud resources. The applications that normally run on-premises—for cost or other reasons—can cloud burst during peak demand to maintain a great user experience.

2. Edge computing

Edge computing allows companies to bring computing closer to their customers or sources of data, so they can better support their remote or branch offices, retail locations, manufacturing plants, or other sites removed from a core datacenter or public cloud. For companies looking for low-latency or disconnected computing, where remote sites can operate without communication with centralized infrastructure, edge computing can help improve infrastructure resilience and application availability. For telco service providers, edge computing can help improve customer experience by pushing applications or content toward the edge tiers in the network hierarchy. They can also enable entirely new classes of services on the edge that require close proximity to customers.

There is a great variety of edge use cases, each with its own unique requirements. The architecture supporting an Internet-of-Things edge characterized by autonomous vehicles will look very different from the edge architecture deployed in a factory or stadium.

Architects need to identify use cases that align with edge computing. If a use case doesn't benefit from reduced latency, real-time monitoring, lower network bandwidth requirements, or other attributes, then edge computing may add complexity without a real business benefit.

[ Also see Datacenter to edge: A reference architecture for managing remote resources. ]

3. Data privacy

Several industries are regulated to protect sensitive customer data (with standards including HIPAA for healthcare and PCI DSS for finance) through secure infrastructure. In many cases, there may also be a requirement to store it within the country of origin. The services offered by a public cloud provider must comply with that industry's regulatory requirements. For example, Google Cloud lists its services that are in scope for PCI DSS.

Architects must evaluate regulatory requirements for their use case and map out the best fit between on-premises and cloud. The hybrid cloud approach allows companies to choose data processing and storage appropriate to data type across on-premises, private clouds, or a suitable region of one or more public clouds.

4. Data protection

Confidential and business-sensitive data needs to be treated differently from less sensitive data. Data with high operational or business value—such as factory operational data or field data from defense forces—may need to be processed, transported, and stored with greater protection. For example, it's probably important to use encryption to safeguard data during all states of its lifecycle, from generation to processing to transmission to storage. The geographic location of private and public clouds could drive the decision about where to process and store sensitive data.

[ Boost security, flexibility, and scale at the edge with Red Hat Enterprise Linux. ]

5. Business continuity

Businesses must continue operating critical functions with degraded or no network connectivity. This approach requires on-premises capabilities. Whether it's a factory, retail store, or hospital, applications and services need to be able to ride out a network outage.

A hybrid cloud can also provide a backup option in case the primary option fails. If the on-premises datacenter fails, a public cloud can be used as a backup; if one public cloud fails, another public cloud can be used as a backup. Taking full advantage of this hybrid cloud use case requires a degree of consistency and portability across the hybrid environment. For example, even big cloud companies can (and do) have massive failures.

6. Innovation

The freedom of choice hybrid cloud offers allows companies to mix and match the best services provided by different cloud providers. For example, they could use AWS Lambda for serverless and Azure AI solutions for manufacturing.

Enterprises using hybrid cloud solutions have found that it speeds business processes, supports collaboration, and delivers cost-effective solutions. The Covid years have provided numerous examples of how the cloud can accelerate digital transformation.

Wrap up

This article discussed the benefits hybrid cloud can bring to various use cases, from telco to manufacturing to retail to transportation. By using a flexible cloud-native approach that extends from on-premises datacenters to public clouds to the edge of the network, enterprises gain the flexibility to deploy solutions that can scale efficiently and cost-effectively.

If you want to learn more about architectural patterns and how various companies are deploying hybrid cloud solutions, check out Red Hat's Portfolio Architecture Center. If you'd like to learn more or have some ideas of your own, reach out to us; we'd love your feedback!