Zero-touch assurance: A tale of two approaches
Service Assurance Automation is a big topic for communication service providers (CSPs) as they manage constantly growing amounts of data in the network. When automation is pushed to its outermost limits, a zero-touch, or a 'lights-out' environment is enabled. Dima Alkin, TEOCO’s Vice President of Service Assurance Solutions, answers questions about zero-touch service assurance.
14 Jul 2021
Zero-touch assurance: A tale of two approaches
Service assurance automation is a big topic these days for communication service providers (CSPs), primarily driven by 5G deployments advancement on the mobile side, and SD-WAN growth on the wireline side of the operators’ business, and the constantly growing amount of data that needs to be managed in the network. When automation is pushed to its outermost limits, a zero-touch, or a 'lights-out' environment is enabled, eliminating human interaction. Dima Alkin, TEOCO’s Vice President of Service Assurance Solutions, answers questions about zero-touch assurance.
How do you define zero-touch when it comes to the area of service assurance?
Zero-touch service assurance is about creating an environment where network failures are automatically identified, assessed, and resolved without human intervention whenever possible, leaving the network operations and engineering teams to focus their efforts on other issues and true outliers. The industry is also moving towards the concept of predictive analytics, which will help by anticipating and preventing network faults before they even occur. Overall, the goal is to minimize network outages and downtime, reduce costs and improve customer satisfaction
What are the most critical areas to focus on for CSPs looking to create a zero-touch service assurance environment?
If you ask five different people to define the most critical areas to automate, you'll get five different answers, and rightfully so. Every operator has their own needs and priorities, which means there is no one-size-fits-all approach to service assurance and network management automation. But we've found that operators typically fall into two broad categories. They tend to generally either take a top-down or a bottom-up approach.
The top-down approach has solid corporate alignment and awareness at the executive management level, which trickles down throughout the organization. Often there will be a significant budget allocated towards automation projects that align with overarching, long-term corporate goals and a well-defined strategy. For instance, an operator may want to fully automate a certain percent of their processes by a specific date or reduce overall network downtime. by some amount.
The bottom-up approach is very different. There typically aren't any overarching corporate objectives at play, which means things tend to bubble up from various departments as individual automation initiatives. Typically, these align around specific departmental performance and cost-saving goals like OPEX savings or reducing how long it takes to resolve trouble tickets. Defining how teams will achieve these goals is not dictated by any corporate directive – but automation is often the best path forward.
What needs to be considered before moving down this path?
Unless you are building a green-field network from scratch, automation will be an evolution. But it's helpful to define goals from the outset, so you can begin planning how to get there. Breaking things down into smaller pieces is how operators can achieve these significant initiatives. One of the first steps is to assess things from an IT perspective. Do we currently have the right tools and capabilities in place to achieve our goals? Understanding this requires asking what should remain and what must be upgraded or modernized.
There is also a middle option; keeping the existing solutions that still deliver good basic functionality and bolting on a new toolset that provides the new features and performance requirements you need to meet your objectives. We've successfully done this with our Helix Analytics solution, which adds a valuable layer of machine learning-based RCA and predictive analytics to any existing service assurance platform.
Once the basic plan is in place and the operator is ready to move forward, understandably, you can't do everything at once. Is it better to automate the low-hanging fruit – or prioritize based on a more strategic long-term vision?
Once again, it depends. Operators with a strategic vision will have a larger plan in place – along with departmental action items and a budget for getting there. For operators who are more hesitant when it comes to automation, there is the need for individual departments to focus on the low-hanging fruit and smaller 'bite-sized' projects. When there's an opportunity to do something much more efficiently, you can typically justify the case. These are often smaller, more manageable projects where you can quickly prove the ROI and show measurable gains. Making these small changes helps prove the use case for additional investments in service assurance automation. Maybe you're not making a dramatic change at an organizational level, but its ripple effects will likely lead to more optimization opportunities down the road.
Network Operations Centers, or NOCs, have been defined as a CSP's central nervous system – a well-staffed, 24 hours a day command and control hub of highly trained network engineers who constantly monitor and maintain the health of the network. Service assurance information feeds into the NOC, along with a myriad of other data sources. Will full automation in this realm ever be a reality – or is it the industry's white whale - something we will chase forever?
Our industry experience shows that this is an achievable goal, but so far mostly in a green-field environment. These are companies that are building a new network from the ground up – like Rakuten in Japan, or in cases where mature operators are essentially building new stand-alone 5G networks and are introducing operational automation from day one. It is much harder to shift from a manual or semi-automated legacy environment to this visionary zero-touch network management approach, especially when you must change and upgrade existing tools and processes while also needing to justify the investment. In our view, full NOC automation is going to be a much more gradual process for mature operators; but that said, there are many “low-hanging fruit” that are ripe for automation within these networks.
What leads to the hesitation towards greater network automation? Is it about the cost, or is there something else at play?
With the focus on building out 5G, many wireless operators globally have spent a fortune on network spectrum. These investments may lead some to hold off on modernizing service assurance and parts of OSS, but they are then left with inefficient systems to manage a cutting-edge new generation network. If you don't modernize the OSS side of the business, it's not sustainable in the long term. At some point, you cannot escape the need for automation in service assurance and other areas. Once you have automation built-in, then you're more efficient in every way. The sooner operators adopt the zero-touch mindset, the better they are in the long term.
How do you define zero-touch when it comes to the area of service assurance?
Zero-touch service assurance is about creating an environment where network failures are automatically identified, assessed, and resolved without human intervention whenever possible, leaving the network operations and engineering teams to focus their efforts on other issues and true outliers. The industry is also moving towards the concept of predictive analytics, which will help by anticipating and preventing network faults before they even occur. Overall, the goal is to minimize network outages and downtime, reduce costs and improve customer satisfaction
What are the most critical areas to focus on for CSPs looking to create a zero-touch service assurance environment?
If you ask five different people to define the most critical areas to automate, you'll get five different answers, and rightfully so. Every operator has their own needs and priorities, which means there is no one-size-fits-all approach to service assurance and network management automation. But we've found that operators typically fall into two broad categories. They tend to generally either take a top-down or a bottom-up approach.
The top-down approach has solid corporate alignment and awareness at the executive management level, which trickles down throughout the organization. Often there will be a significant budget allocated towards automation projects that align with overarching, long-term corporate goals and a well-defined strategy. For instance, an operator may want to fully automate a certain percent of their processes by a specific date or reduce overall network downtime. by some amount.
The bottom-up approach is very different. There typically aren't any overarching corporate objectives at play, which means things tend to bubble up from various departments as individual automation initiatives. Typically, these align around specific departmental performance and cost-saving goals like OPEX savings or reducing how long it takes to resolve trouble tickets. Defining how teams will achieve these goals is not dictated by any corporate directive – but automation is often the best path forward.
What needs to be considered before moving down this path?
Unless you are building a green-field network from scratch, automation will be an evolution. But it's helpful to define goals from the outset, so you can begin planning how to get there. Breaking things down into smaller pieces is how operators can achieve these significant initiatives. One of the first steps is to assess things from an IT perspective. Do we currently have the right tools and capabilities in place to achieve our goals? Understanding this requires asking what should remain and what must be upgraded or modernized.
There is also a middle option; keeping the existing solutions that still deliver good basic functionality and bolting on a new toolset that provides the new features and performance requirements you need to meet your objectives. We've successfully done this with our Helix Analytics solution, which adds a valuable layer of machine learning-based RCA and predictive analytics to any existing service assurance platform.
Once the basic plan is in place and the operator is ready to move forward, understandably, you can't do everything at once. Is it better to automate the low-hanging fruit – or prioritize based on a more strategic long-term vision?
Once again, it depends. Operators with a strategic vision will have a larger plan in place – along with departmental action items and a budget for getting there. For operators who are more hesitant when it comes to automation, there is the need for individual departments to focus on the low-hanging fruit and smaller 'bite-sized' projects. When there's an opportunity to do something much more efficiently, you can typically justify the case. These are often smaller, more manageable projects where you can quickly prove the ROI and show measurable gains. Making these small changes helps prove the use case for additional investments in service assurance automation. Maybe you're not making a dramatic change at an organizational level, but its ripple effects will likely lead to more optimization opportunities down the road.
Network Operations Centers, or NOCs, have been defined as a CSP's central nervous system – a well-staffed, 24 hours a day command and control hub of highly trained network engineers who constantly monitor and maintain the health of the network. Service assurance information feeds into the NOC, along with a myriad of other data sources. Will full automation in this realm ever be a reality – or is it the industry's white whale - something we will chase forever?
Our industry experience shows that this is an achievable goal, but so far mostly in a green-field environment. These are companies that are building a new network from the ground up – like Rakuten in Japan, or in cases where mature operators are essentially building new stand-alone 5G networks and are introducing operational automation from day one. It is much harder to shift from a manual or semi-automated legacy environment to this visionary zero-touch network management approach, especially when you must change and upgrade existing tools and processes while also needing to justify the investment. In our view, full NOC automation is going to be a much more gradual process for mature operators; but that said, there are many “low-hanging fruit” that are ripe for automation within these networks.
What leads to the hesitation towards greater network automation? Is it about the cost, or is there something else at play?
With the focus on building out 5G, many wireless operators globally have spent a fortune on network spectrum. These investments may lead some to hold off on modernizing service assurance and parts of OSS, but they are then left with inefficient systems to manage a cutting-edge new generation network. If you don't modernize the OSS side of the business, it's not sustainable in the long term. At some point, you cannot escape the need for automation in service assurance and other areas. Once you have automation built-in, then you're more efficient in every way. The sooner operators adopt the zero-touch mindset, the better they are in the long term.
