The cloud has a lot going for it, but it’s not infallible as many people and enterprises realized again last week when Amazon Web Services (AWS) suffered a prolonged outage that impacted both basic applications and colossal businesses with hundreds of millions of customers.

When the world’s largest public cloud fails, major organizations that rely on it are left in limbo. 

Amazon’s outage last week, which lasted the better part of a day, serves as yet another reminder that telecom networks can’t yet, and may never be able to, fully rely on cloud infrastructure for critical services and workloads.

The benefits of public cloud are clear — efficiency, scalability, and the ability to consolidate functions with less equipment. For telco operators this translates to better economics, business agility, and accelerated innovation at the pace of software, explained Don Alusha, senior analyst on telco cloud platforms at ABI Research.

Telecom Requires Strict Uptime Guarantees

However, “communications service providers have much stricter service level agreements (SLA) and service-uptime guarantees remain stringent relative to cloud software and platforms,” he wrote in response to questions. 

“Operators stand to benefit from cloud innovation and agility, but they need to balance the risk of deploying cloud-driven methodologies and tool chains,” Alusha added.

These risks grow in line with the number of services, and the critical nature of those respective services, deployed by carriers in the cloud. 

Mobile network operators that run back-end systems such as operations and business support systems, provisioning, and billing in the cloud could be impacted by outages, according to Sid Nag, VP of cloud services and technologies at Gartner.

Most mobile operators don’t run core connectivity functions in the cloud today, he added, but this is changing in the 5G era as carriers embrace virtualization and the automation of network maintenance and operations. 

Public Cloud Telco Role Remains Unclear

“The exact role that public clouds will play in how telcos position themselves in the impending competitive landscape remains to be seen,” Alusha said. This uncertainty extends to the level at which cloud-based software programs can and will satisfy the strict requirements of telecom architecture, he added. 

“It may well be that there is a need for further maturity and a hardening process that guarantees stringent, telco-grade SLAs, and service-uptime agreements for critical network connectivity when using public clouds,” Alusha explained. 

Appledore Research founder and Principal Analyst Patrick Kelly agrees, adding that while cloud infrastructure provides huge economic benefits, “the resiliency of the cloud for critical infrastructure must still be hardened in the areas of security and reliability.”

This risk is not specific to hyperscalers like AWS, as mobile networks already suffer outages completely unrelated to the cloud.

Outages Not Exclusive to Hyperscalers

“Outages happen” said Chris Antlitz, principal analyst at Technology Business Research. “They can happen in the public cloud, and there are many historical examples where it has happened in public clouds. Telcos that run their own private clouds are not immune to outages.” 

The recent AWS outage won’t impact the secular trend of information and communications technology workloads moving into the cloud, he said, adding that outages also provide hyperscalers an opportunity to learn and get better.

“We would expect the resiliency and reliability of public clouds to continue to improve over time,” Antlitz said. 

While the cloud, like any technology going through its first wave of adoption by wireless network operators, presents many risks and unknowns relative to performance, carriers that overcome these challenges early can gain a competitive advantage, according to Alusha.

Mass market services such as voice and data shouldn’t encounter atypical issues due to the cloud’s growing role in network infrastructure, but customized services enabled by the edge and private networks are a bigger challenge, Kelly explained. 

“Critical latency-sensitive applications are more important and could expose providers and suppliers in the value chain to increased liability claims that extend far beyond connectivity SLAs,” he said. 

Carriers Confront Cloud Challenges

Mobile network operators are accustomed to confronting risks and that experience could play in their favor as telco cloud services eat further into legacy infrastructure.

Carriers should evaluate different cloud strategies — multi- and hybrid cloud, for example — in line with their market positioning and growth strategy, Alusha explained. “Operators should hedge their bets in alternative and competing cloud platforms to change the structure of current systems and processes to produce more of what is desirable and less of that which is undesirable,” he added.

This puts the onus on enterprises and telcos to build and maintain relationships with multiple cloud providers and suppliers, according to Kelly. Moreover, he said, “technology like network slicing helps isolate impacts and provide redundancy.”

Carrier technology heads likely took note of AWS’ outage last week, but they also just as likely took comfort in the fact that mobile networks are shifting to the cloud much slower and with the amount of constraint required of critical infrastructure. Many of those reservations will endure until public clouds can guarantee 99.999% reliability.