How AI solutions can bring real value to telecom infrastructure

[Claus_Pedersen]

Almost every industry has been investigating how to use artificial intelligence (AI) to increase revenue, reduce costs, or enhance customer experience. Many organizations recognize that AI is critical to their business strategies, but they often struggle to determine how it fits into their current operations. Telecom stands to be one of the industries than can benefit from AI.

So — Where is the best place for telecom leaders to start implementing AI solutions?

One aspect that's often discussed is AI's effect on the excellent customer service experience. You, as a telecom service provider, may already be using virtual assistants that interact with customers to improve customer engagement, or systems that automatically route customer requests based on qualifying questions that ensure prospects enjoy a positive and satisfying outcome to their inquiry.

However, these applications don't tell the full story of how AI can aid telecom service providers. Some of the biggest benefits can be realized behind the scenes—with AI monitoring your IT infrastructure for predictive maintenance.

Telco AI solutions for predictive maintenance

The changing face of the telecom industry means many service providers are seeing demanding challenges being placed on their network infrastructure. Usage is growing, as are demands for a quality experience. At the same time, networks have to improve capacity and reliability to meet the challenges of the modern mobile economy—including the internet of things and the commercial evolution to 5G—as deployments proliferate during 2019.

All this needs to be achieved while keeping the costs low.

Telco service providers are responding to these challenges with new technology innovations such as Network Function Virtualization (NFV) where telco applications are deployed on commercially available off-the-shelf telco ready x86 IT infrastructure. Also, they are moving intelligence to the edge using new architecture concepts called multi-access edge computing (MEC).

This offers a great deal of flexibility, but it can also introduce further security and operational challenges due to the increased demands of maintaining and managing a sophisticated combination of physical and virtualized distributed systems.

This is one area where AI can play a vital role in your success. Much effort is often wasted by ignoring telco IT infrastructure maintenance problems as they are mounting, or discovering them after something has already gone wrong. And large sums of money is wasted on preventable remediation.

This break-fix approach is a risky tack for telecom service providers aiming to deliver consistent service levels. It's rarely cost-effective. At the very least, there would be a hiatus until things get fixed, and at worst, an unpredicted failure in telecom infrastructure could inflict long-lasting downtime that does irreparable damage to the company's reputation.

AI can monitor telco IT infrastructure and provide administrators with a more intelligent and automated approach to maintenance. The amount of potentially valuable data that can be gathered from distributed networks might be too overwhelming for even the best network managers, but it is vital fuel for a machine-learning algorithm.

Using AI for actionable insight

HPE InfoSight puts the focus on prevention. The AI-powered predictive analytics platform, newly available for HPE Infrastructure, looks across the entire infrastructure—from edge to core—to take the guesswork out of management and fault-finding.

Hundreds of predictive signatures—all revealing information about availability, performance, security, capacity, data protection, and more—are continuously monitored in every system across the installed base. For example, if HPE InfoSight uncovers a problem, it resolves the issue and prevents other global telco systems from experiencing the same problem. First and second level support are covered, so escalations are no longer necessary. In the rare instance you need it, InfoSight enables direct action for level-three alerts.

HPE InfoSight already has comprehensive information about the telco IT infrastructure environment, so even the most complex issues can be resolved quickly. If a problem is detected that can't be automatically addressed, then support engineers can investigate and reach out with the right fix.

The security of customer-originated data is a top priority for HPE. At no time is customer data sent to HPE InfoSight. Data collection by HPE InfoSight is strictly limited to configuration-related and performance-related data. Depending on the type of data, customers can choose to opt in or out of providing specific datasets to HPE InfoSight.

Delivering carrier-grade services can demand a lot of tuning. If this tuning is entirely manual, it can result in a lot of unrewarding effort. HPE InfoSight takes the guesswork out of managing telco IT infrastructure. It identifies how a telco can improve performance, optimize resources, and plan for the future. Even in complex, NFV virtualized environments, its cross-stack analytics pinpoints abnormal performance issues between servers, storage, and virtual network functions (VNFs) and recommends the right remediation through app- and resource-centric modeling, all in one click.

New services, real solutions

New technologies are often disparaged as solutions in search of a problem. But problems in telecom IT infrastructure are real, and potentially fatal flaws. Delivering carrier-grade services is a challenge in the best of times, but the popularity of new services, coupled with increasing adoption and user expectations, means that telecom service providers all need to up their game, and be prepared for issues that could arise.

AI solutions do not have to be relegated to futuristic or exotic applications to benefit business. They are most effective when applied to the mundane but essential act of keeping the lights on. Additionally, solutions such as HPE InfoSight can help telecommunications service providers improve their offerings, and grow their business.