The NetOps Struggle is Real
NetOps practices are lauded for their ability to create a stable, transparent, and flexible network infrastructure. However, this agile and collaborative strategy is not without its pitfalls.
Network Operations (NetOps) is a popular strategy that takes cues from DevOps principles and applies it to network management tasks. The goal is to create a network that is more flexible and agile from a network moves/adds/changes perspective while continuing to maintain high levels of reliability. Within this movement are several new and shiny network management tools designed to aid NetOps staff with virtualization, automation, and monitoring tasks. But despite IT management buy-in and a host of useful tools, plenty of organizations have tried and failed to implement NetOps in a meaningful way. Here are four common reasons why.
1) NetOps principles are counter to what many IT administrators were originally taught about network management
For decades, network administrators were taught the old “if it ain’t broke, don’t fix it” approach to production network changes. At that time, the network served the simple function of pushing frames and packets from one location to another. There were very few ways to adjust a network from an application performance perspective. Thus, any configuration modification that was deemed necessary was overly scrutinized so that outages due to human error would be limited as much as possible.
NetOps practices, on the other hand, use advanced AI, machine learning, and automation to reduce configuration-related downtime. This allows more changes to be made in rapid succession with little change in risk. However, keep in mind that the original mindset is likely still engrained in seasoned administrators. It often takes a considerable amount of time for these admins to grow comfortable with the speed at which NetOps must now function.
2) A lack of NetOps tool training
It shouldn’t be downplayed that modern NetOps tools are complex platforms that use advanced AI and machine learning techniques that require fine-tuning from time to time. Not understanding how to interpret and calibrate these tools can create situations where operations staff loses visibility into the performance of portions of their network.
If not properly implemented and managed, NetOps tools can create visibility gaps, act on false positives and send administrators down the wrong path to performance gains. This can lead to situations where tools no longer become trusted and end up being more of a hindrance than a benefit.
3) Focusing on the wrong metrics
Key performance indicators (KPIs) are metrics used to gauge the success of a business goal. Traditional network KPIs largely focused on performance and reliability. While important, NetOps adds additional indicators that must be accurately tracked and monitored over time. Examples include:
Overall number of automated network tasks
Mean time to network service delivery
Speed of onboarding/offboarding network users
Ratio of manual network changes vs. automated changes
Time to failure detection and root cause analysis (RCA)
Network configuration self-service portal usage
Without analyzing these types of KPIs, it becomes impossible to know whether existing NetOps practices are working or what adjustments need to be made.
4) Lack of cross-team collaboration
NetOps isn’t simply about changing practices and philosophies within the network team itself. Instead, the approach crosses multiple IT department silos to include application, server, and software development teams. While network agility benefits these non-network-centric teams by providing them with faster network configuration changes that is far faster than traditionally expected, collaboration between teams is an absolute must.
The two most common NetOps cross-team collaboration breakdowns include a lack of inter-team communication and mishandling of shared tools. The lack of communication issue largely boils down to network changes that were made by one team but not communicated to others through centralized channels such as service desk ticketing systems. A mishandling of shared tools often occurs when teams learn how to make automated configuration changes using different procedures. This leads to network configurations that are not uniform – and could ultimately create confusion between groups.
NetOps, is it worth it?
Simply put, yes, it is worth overcoming these types of common challenges to achieve the level of agility that NetOps principles offer. While these types of IT strategies remain a challenge due to cultural and technical deficiencies, it certainly isn’t an impossible task. Given plenty of time, the right tools, and a proper mindset, anyone can overcome NetOps struggles.
Read more about:
Network ComputingAbout the Authors
You May Also Like