Not all private wireless solutions are created equal
Given the very tight standards that 3GPP sets for 4.9G/LTE or 5G, it might be easy to assume that all solutions conforming to those standards are roughly equal when it comes to providing highly available and reliable wireless connectivity. But, as they say, the devil is in the details. Conforming to the standard is only the beginning. How the components and solution architecture are executed can make for very unequal results.
In this two-blog series, I want to address this question directly, first focusing on availability and, in the second blog, addressing reliability. Nokia, through its extensive work in private wireless since 2012, has gained a lot of learning in this space, including what does not work. We have had a chance to see a fair amount of our competitors failing miserably at that, based on a lack of understanding of the many aspects and factors that makes a network worthy of powering use cases, where a person’s life or the surrounding environment may be in danger in case the network fails.
High availability in the range of, or above, 99.95% is typically expected for systems supporting mission- and business-critical use cases, and it is something that many of the leading private wireless solutions can support. If well designed for the site where they are deployed, private wireless that uses 3GPP standard technologies, can meet or even exceed such requirements. Typical features of these systems include multiple layers of redundancy in the core (hardware and geographic), overlapping coverage of cells, redundant backhaul and transport, hot-software updates, and graceful redundant server switch-over. For edge-based solutions, they also require the ability to support continuous operation, even if losing the WAN connectivity to the cloud or central office location.
When more stringent requirements are required, our private wireless solution capabilities and enterprise design practices at Nokia can beef up the solution by using multiple radio spectrum bands to create multi-layer radio connectivity, geo-spatial separation of redundant small cells, and a third level of redundancy from a public network core or a centralized core at HQ, for example. Today, our Nokia Digital Automation Cloud (DAC) solution, gives us real-time benchmarks on how the customers’ networks we are managing are operating. On average we have achieved 99.999% uptime in the last 3 months (including planned but also unplanned downtime). For the most stringent deployments, using either DAC or our Modular Private Wireless solution (MPW) and looking at the hundreds of networks we are operating, we are exceeding 5x9s.
In other words, the missing ingredient in achieving very high availability is the experience of delivering end-to-end solutions for decades, whether private wireless or other critical enterprise networks such as GSM-R, IP-MPLS, microwave and optical. Experience and the right feature set for critical networks are the key ingredients to providing networks that our customers can rely on.
The impact of critical network experience
At Nokia, we pride ourselves on being the vendor with the longest experience both in private wireless and running critical networks. We have been delivering mission-critical GSM-R for railway operators for 30 years. And in private wireless, we pioneered the technology and have been leading the industry for nearly 10 years.
This experience means that we understand the requirements of the industry. We know how different a network is in an underground mine versus a factory and how the requirements of various use cases can differ widely. Our many years running critical networks in our focus 12 industrial and government and cities segments have been essential to improving our software across all elements of the solution and ensuring the code is as sturdy as it can be.
In this regard, our deep experience in the telecom sector is also a huge benefit. Many of the components and software that we use in our private wireless solutions are also running on the nationwide public networks of our several hundred mobile operator customers operating around the world. Due to the sheer size of public networks, we must ensure reliability for networks with anywhere from 10 to 100 million subscribers in some of the harshest environments in the world, including deep jungle, sub-Saharan deserts and arctic zones in the north and the south — not to mention the most difficult environment of all, dense metropolitan urban centers. These networks provide mobile connectivity for ever-more-demanding subscribers, but as well, the critical services running on these networks such as public safety and logistics. We know intimately what telco-grade reliability means, and we are leveraging the techniques used there for our private wireless customers, including the continuous software improvements since the early days of 2G, which means our software-code is also well-hardened.
End-to-end is key to total system reliability. It starts with individual elements, how they are assembled and, very importantly, how they all work together. Enterprise customers understand this as they rarely buy operational technology (OT) elements piecemeal for their more complex systems. They historically have relied on trusted system integrators to provide them complete machines or production lines.
At Nokia, we take the same approach. Even if some of our solutions are modular and can operate with other supplier’s elements, most enterprise prefers to trust a single supplier for their private wireless networks, and for good reason. Nokia private wireless solutions are all pre-integrated and tested to function well together, end to end. This is more impressive when you understand that we have one of the widest portfolios of key elements for private wireless. We provide elements ranging from industrial devices to radios, the backhaul and fronthaul network, the core and edge cloud (including the server hardware), IP-MPLS routing, microwave and optical links, plus operations and maintenance and, finally, the applications running on the network. These are all pre-tested to work like a swiss-clock. In comparison, many of our competitors would need four to five partners just to assemble such a solution or, in other words, more chances for glitches, requirements for regression testing at every new software release (= downtime) and, most importantly, the chance for multiple suppliers to pass the blame when things go wrong.
Still, even Nokia needs trusted partners to address every single use case in the market. But the point is that we need less than many others, and we take a systemic approach that we call “Segment Blueprints” to designing private wireless solutions that integrate, for example, specific industrial partners’ elements, software, etc. These blueprints go beyond our end-to-end solution and are created based on our learnings from deployments that we have had in each of the sub-segment we focus on. We create pre-integrated and tested solutions that can tackle the key use cases of each segment with a range of preferred partners assets. These systems go through our Enterprise services labs around the world to be tested and validated to reduce the need for customization and remove unwelcome surprises during deployment and operations.
In conclusion, the availability of a private wireless solution is not something to take for granted when choosing a vendor. Experience in designing and running critical network is essential. In the next blog, I will discuss how system architecture and, most importantly, the quality of the radio should not be underestimated to meet the very high reliability expectations and requirements needed for Industry 4.0 applications.