By Dwarakanath Yadavalli, Software Engineering Manager, Managability, Juniper Networks IEC
A digital enterprise is looking to deploy its new data center comprising of tens of thousands of devices. A prominent telecommunication service provider is seeking to provide a very personalized experience to its customers. Another global enterprise is looking to scuttle a globally distributed denial of service (DOS) attack on its information technology (IT) infrastructure.These aren’t one-off isolated incidents but instead are routine activities in the digital world.
Some of the contributing factors that make these activities frequent are:
– Never ending gravy train of technology and social innovation
– Increasing digital presence
– Expanding digital lives of users
– Omnichannel strategies of enterprises
– Hyper-competition in several industries
– Ransomware attacks and industrial espionages that are driven by the dark digital underworld
These factors further possess substantial interplay.
A simple Google search on “pace of innovation” displays tons of charts, graphs and point-of-view papers that emphasize and underscore that it is a train that is never slowing down to let anyone on and that the end-user demand will never be satiated. That, of course, is needed to keep pace with the changing needs and tastes of consumers.
Across several industries leadership is shaped by the digital might – Amazon in online retail, Uber in ride-hailing, Airbnb in hospitality, LinkedIn in recruitment, PayTM in online payments. These have all found success through their investments in IT-driven front-end and back-end.The net effect of this reliance on IT is the increasing size of data centers built to crunch humungous amounts of data in near real-time to glean insights about customers and competition among several other workloads.
So, where does automation apply?
A critical component of securing the competitive advantage is an investment in continuous improvement and automation of operational processes. Given the current state of the networking industry, networks could derive huge benefit from investments in automation.
Automation is no longer an afterthought; it is needed, beginning with network deployment to all the way to routine monitoring post-deployment and for everything in between. If one were to add the business requirements of productivity, operational expenditure control, profitability and time-to-market/deploy then Automation becomes one of the top priorities for any serious customer and vendor too. Automation provides new technologies for managing networks and abilities to createcomplexity.In chasing the next big thing, application developers, service providers, and other organizations are rolling out new services requiring back-end systems that are getting more and more complex by the day.
Consider these near future use-cases of self-driving cars that require ultra-low latency, an industrial Internet that predicts machine failure and automatically assigns fixes, or even just advances on the mobile network like 5G connectivity, network automation is going to start cropping up in whole new ways.
Levels of Automation
There are several Levels of Automation (LOA) taxonomies proposed over time. These levels range from extremes of no/low automation, where the human performs the task manually to complete automation wherein the computer is fully autonomous. All automated systems operate at some level within a given scale.
Inspired by the way humans process information, Parasuraman et al, defined a simple four-stage model describing system functions that process information:
– Information Acquisition
– Information Analysis
– Decision and Action Selection
– Action Implementation
These stages of automation (SOA) can be implemented to different degrees.Considered together, each task in a network operation can be at a different level and stage of automation, resulting in the automation continuum.
Challenges and Opportunities
As new challenges emerge, the digital enterprises will have to consider the following innovations — and challenges — for automation to succeed.
Human capital reallocation: With automation comes a perceived threat to job security. However, network automation isn’t going to replace anyone; humans are still a necessary part of the equation. Rather, it’s going to augment them while freeing up the time to focus on more critical tasks.
Human intervention is needed to provide the intent in a given network; however, that is rapidly changing as intent-based networking catches on. The right automation tooling integrated with an existing monitoring system can enhance the quality, speed, and accuracy of incident management; be it networking or in general.
Re-training staff to improve their provisioning prowess is a must in the automated future, shifting from manual to automatic; configured to coded; change-controlled to continuous; and from tracing and rummaging through minutiae to drilling down through layers of abstraction.
Automation as a facilitator: As data crunching moves, from devices to the distributed cloud at the edge, monitoring application performance must also change to ensure real-time processing for seamless low-latency applications. Virtualization brings the ability to flexibly cater to many different end-users’ needs, but a human intervention to maintain all SLAs is nearly impossible at scale. Only automation can enable zero unplanned downtime across the spectrum.
Machine learning and AI: Intelligent software using machine learning and artificial intelligence will respond to issues and fix them in real time — to enable network operators to better manage applications and services that rely on their networks. Machine learning and AI will help in detecting anomalies using predictive analytics to correlate and combine data based on demographic, geographic and traffic patterns.
Automation Roadmap: Several companies are in the early stages of replacing manual tasks and operations with automated systems. Automating the whole network could take time and probably will happen in parts. Enterprises need to layout a roadmap for their automation journey, as success in automation requires a concerted and sustained effort; not a patchwork of disconnected responses after the fact.
Automation may start with a linear approach of running a set of instructions executed until the end. It may then evolve into an orchestration of multiple such linear instructions across devices. However, the end goal is to have a closed loop of automation driven by an all-pervading telemetry infrastructure and data thus obtained, funneled into intelligent and autonomous decision making.
Automated networks will pave the way for the future where applications can self-assemble.
Ultimately, organizations must make sure that their network becomes self-learning and automated to ensure that they thrive in the digital age.