The Cloud-native computing for telcos: definitions, challenges and opportunities

EXECUTIVE SUMMARY

Mastery of cloud-native computing is critical for DSPs, but adoption is fraught with challenges

Telco industry players should prioritise cloud-native computing if they are committed to becoming DSPs

Key recommendations

IS BECOMING A DSP THE RIGHT STRATEGY?

Not all CSPs are willing or able to transform themselves into DSPs

DSPs expect to gain significant business benefits over other types of telco, but need a strong appetite for risk

UNDERSTANDING CLOUD-NATIVE COMPUTING

The principles of cloud-native computing have developed over the past 5 years and its definition is still evolving

Microservices oriented: Applications need to be developed as, or refactored into, microservices

Containerised: Applications must be packaged in containers

Dynamically orchestrated: Application containers must be centrally and dynamically orchestrated

ACCELERATING CLOUD-NATIVE ADOPTION

True believers must do more to accelerate their own rate of adoption, while hybrids have a vested interest in accelerating that of their vendors

Key telco barriers to cloud-native computing must be understood and addressed before mainstream adoption can take place

Selecting the right base technologies is tricky in a fragmented ecosystem, but the current winners are emerging

VENDOR SNAPSHOTS

Canonical

Cloud Native Computing Foundation (CNCF)

Docker

Mesos

Rancher

Red Hat

VMware

ABOUT THE AUTHORS AND ANALYSYS MASON

About the authors

Analysys Mason’s consulting and research are uniquely positioned

Research from Analysys Mason

Consulting from Analysys Mason

List of figures

Figure 1: Drivers of digital transformation and recommendations for CSPs

Figure 2: Benchmarks for cloud-native computing KPIs set by FANG companies

Figure 3: Three ways in which CSPs should advance cloud-native network function virtualisation

Figure 4: Characteristics of the three types of CSP

Figure 5: KPIs for the three types of CSP

Figure 6: The evolution of cloud-native computing

Figure 7: The CNCF’s definition of microservices oriented

Figure 8: Example of how a virtual network function might be decomposed into microservices

Figure 9: CNCF’s definition of container-packaged applications

Figure 10: Comparison of containers and virtual machines

Figure 11: CNCF’s definition of dynamically orchestrated

Figure 12: Overview of the container platform and container engine

Figure 13: Strategies for accelerating cloud-native adoption

Figure 14: Four key barriers to rapid adoption of cloud-native computing

Figure 15: Overview of the CNCF tooling taxonomy