Use cases such as autonomous vehicles, drone management, energy networks and advanced factory robotics will all demand the intelligent handling of massive amounts of data for smart decision making at the edge of the network as well as in devices, right where it is needed most.
Over the next 12-18 months, 5G adoption will pick up speed, achieving far greater proliferation across North America, Europe and parts of Asia. 5G will be heavily dependent on edge computing to meet the low latency requirements of the next generation of use cases.
These significant developments will put greater emphasis on the need for distributed cloud infrastructure, with edge solutions and cloud deployment accelerating across enterprise environments. Coupled with a global transition to 5G, distributed cloud will emerge as the preferred infrastructure model, driven by increased demand for flexibility, reduced latency and support of a remote workforce.
However, the development and deployment of intelligent edge systems comes with its own set of old and new world challenges.
The old vs the new
We are in the midst of an age of disruption. Emerging technology trends, constantly evolving customer demands, new industry dynamics, and business models are setting a precedent for what is expected of developers, both new and old, in years to come. Everything is expected to be autonomous with critical data processed in real time. Cloud-native trends driven by AI, 5G, and intelligent systems at the edge, are leading a surge in demand for developers with new skill sets.
Meanwhile, legacy code continues to run in embedded systems for those mission critical sectors – industrial, aerospace & defence, transportation, utilities – as it will take time to further modernise those systems with greater intelligence and automation. Layered on top of this, is a shift in the developer workforce from boomer, to millennial, and beyond.
As the current generation of developers, who understand legacy code running in older languages, start to retire, we face a situation where colleges and universities have not focused on training the next generation of developers to work with these legacy codebases. This will present a
significant hurdle for enterprises looking to engineer greater flexibility into the fabric of their business and meet with new demands.
Enterprises will need to act quickly and invest in the appropriate training and tools to support this workforce transition. While this may seem like unchartered waters during challenging times, actually what it truly creates is an opportunity to realign the future of intelligent systems at the edge around the newer cloud-native skills and thinking that recent graduates employ.
Training the winners
This new world driven by intelligent edge systems requires a range of new skills, thinking, and knowledge. Knowing where the data is coming from primarily, whether edge or cloud, will shape the way teams should be organised and what expertise is most critical.
A deep understanding of mission critical intelligent systems including AI & ML, real-time analytics, security and network reliability are all fundamental to realising an edge computing system or an embedded system for edge computing.
With enormous quantities of data to process and analyse in real time in an edge environment, developers must leverage automation, AI and ML capabilities to better understand and use the data at hand. This will see developers understand not only the error, or the occurrence of a fault, but the detail of what happened in the run up to the fault – vastly reducing the chances of incurring that fault again.
Engaging with real time analytics will see developers empowered to anticipate, identify and resolve faults or errors as or even before they happen.
Development teams will need to understand a breadth of coding languages, and be well versed in data science, before reaching this promised land of AI and ML infused systems. Equally the teams will need to develop understanding for application deployment, maintenance and security from cloud to edge. Adaptability will also be key for developers, enabling them to update and reprogram systems and software in newer more flexible environments. Modern developers should be encouraged to design with ambiguity in a modular way, for systems to be more adaptable, deriving benefits and value from real time insights.
DevSecOps will continue to play an instrumental role in the development of systems that breed a wealth of new use cases and applications at the network edge. Security must always remain front of mind as the attack surface grows along with a more distributed environment, securing the DevSecOps environment and providing the tools to ensure that the output is as secure as possible. DevSecOps will ensure customers receive the products they demand in the fastest time possible, therefore it is critical developers work harmoniously with the rest of the business to fully understand customer and industry challenges. However, this is all theoretically unachievable if the skills gap is not addressed.
Crossing the finish line
As edge computing systems continue to become more ubiquitous, developers don’t only need to up-skill, they must also be mindful of broader customer challenges. In an increasingly software defined world, developing high quality software is a top priority. Edge computing systems today pack many features that are all differentiated through the software. The expectation is that more features are delivered in less time and at lower cost without compromising the quality of the software. The need for continuous deployment means the entire process needs to be more automated and reliable.
Software culture and best practices are rapidly evolving. The practices of adopting Agile and DevOps for software developments are an inroad into edge computing systems. It demands the usage of modern software development tools, techniques and programming languages. Developers are expected to familiarize themselves with modern tools without impacting customers’ productivity and efficiencies, regardless of the learning curve they’re faced with.
Cybersecurity must also be at the forefront of the development and operations of these systems. With edge computing comes connected smart devices with different levels of connectivity and deployment scenarios. This increased attack surface requires an integrated secure development process, as well as security rich features in the products that provide defence during deployment and thereafter, once in autonomous operation. The final challenge customers face is handling data at scale, they require that information to be converted into actionable insight and failure to do so comes at the risk of security and safety of enterprise personnel.
The shift to the edge is occurring rapidly, spurred on by a greater impetus around distributed infrastructure and a desire to realise the potential of the intelligent edge. The advent of 5G brings promise closer to reality for edge computing environments, and enterprise intelligent systems demand security, safety and reliability.
Legacy code and current developer skillsets are not enough to achieve this. Developers will need a deep understanding of the new technologies that enable the automation and orchestration of a new generation of use cases at the network edge. 5G and the edge will propel enterprises into a fully digital, AI driven future. Developers with crucial new skillsets, adopting a DevSecOps approach, will be the driving force that puts enterprises into pole position as edge leaders.
Author details: Matt Jones is Vice President, Global Engineering, Wind River