By Bernard Theot
Technical Director Metrology
Supervision, Monitoring Vs Metrology - Definition
To begin the subject, we will look at the definition of three terms that are often confused.
First of all, let's talk about "Monitoring". This term refers to the act of " monitoring ". Supervision is one of its components, rather real-time oriented. Monitoring " means knowing the current state of something, but also the history of its past states, by means of values and figures. This is where we find the distinction between the two notions of Supervision and Metrology.
In the term Metrology, we find the root " metre/metric ". It is the discipline of obtaining, keeping and plotting the numerical value of a load. There are many examples, such as: the percentage of CPU used on a server, the number of people connected to a website, the traffic coming in and out of a switch port. In many cases, Metrics can simply be used to plot graphs, which are well known in the field of monitoring.
The field of metrology
To address the ever-increasing complexity of application ecosystems, Metrology is a constantly evolving field. It is able to respond to a wide variety of audiences, such as network engineers, developers, testers, as well as the various business lines.
In order to address these different types of audiences and to cover all the components of the IT ecosystem, Metrology is organised in three interdependent dimensions: The User, the Application and the Network. Different technologies are used to instrument these dimensions. Each of these technologies is specialised in the domain it covers. A correlation is then set up to present the sum of the Metrology results in a unified visualisation portal.
We will now go through each of the dimensions of Metrology.
The IT user - User Centric
The EMU not only monitors the resources with which the employee interacts, but also allows for proactive automated remediation to continuously improve the user experience. 
The application - Application Centric
The Application dimension is addressed byApplication Performance Management (APM) technology. This technology enables IT teams to ensure consistent availability and a high level of performance for the company's applications. Websites, mobile and business applications are all candidates for monitoring. In today's highly connected digital world, monitoring extends to services, processes, hosts, logs, networks and, of course, end users; whether external customers, partners or employees.
Every day, the user accesses applications to shop online, watch TV shows, connect to social networks, manage finances and work. The user is frustrated when they experience application crashes or when the application responds slowly.
There are many variables that can impact application performance: code errors or software bugs, database slowdowns, network or host performance degradations, media types and operating systems, etc.
Websites, mobile and business applications are increasingly complex; built on millions of lines of code, hundreds of interconnected digital services and hosted in multiple cloud services. For these reasons, digital teams are using APM solutions that allow them to understand and troubleshoot the various components that can impact performance.
Example of application interactions identified by an APM solution
Real-time status monitoring of several hundred application services.
The Network - Network Centric
The network dimension is addressed by NPM(Network Performance Monitoring) technology. This technology is a set of methods to monitor and ensure the best performance levels of a network. The key performance metrics typically collected by NPM probes are : Error rates, delays, packet losses, retransmissions and bandwidth usage.
NPM solutions enable a proactive approach to identifying and reducing bottlenecks and other network incidents. These problems affect not only end users, but also overall business operations, including basic maintenance tasks.
Today's increasingly complex IT infrastructures require relevant information about end-to-end visibility. Today's network cells deploy a significant number of components to improve productivity, increase network performance and ensure the best user experience.
Maintaining this level of visibility is a real challenge in the face of technology stacking, both in third-party environments hosting SaaS applications and in the face of increasing internal application density.
The NPM identifies incidents related to internal functions by evaluating the performance metrics of each component of the network infrastructure. Real-time monitoring of performance metrics or logs allows IT teams to proactively resolve incidents by identifying potential causes of outages before they impact production.
Security threats such asmalware are often the cause of poor network performance and can quickly spell disaster for the business. An NPM solution must be able to detect insecure network devices, malware and other network vulnerabilities. This allows IT departments to develop network performance and fault management protocols to manage and prevent weaknesses before confidential data or end users are affected.
Assessment of the level of maturity
In order to allow the company to position itself and monitor its progress in Metrics, Gartner has created a maturity scale. Depending on the processes already established, IT can situate its current level of maturity before moving on to the next level by adopting the related processes. Without being obliged to reach the maximum level of value, IT has the choice to plan its progression to the desired level.
Current developments in Metrology
As a result of the increasing adoption of Metrology by the different actors in the company, many developments have started. Some of them are already very elaborate. We will soon propose other articles that will allow us to detail the most relevant ones.
One of the most interesting areas is Observability. By feeding on metrics and data collected, for the most part via Metrology, Observability goes far beyond simple error detection and really looks for the causes of malfunctions. Observability helps modern teams to :
- Delivering high quality software on a large scale.
- Develop a culture of sustainable innovation.
- Optimise investments in cloud and modern tools.
- See the performance of their digital business in real time.
A second important area is DevOps. This discipline combines Development(Dev) and Operations(Ops) to bring together people, processes and technologies to continuously deliver value to customers.
DevOps enables the coordination and collaboration of previously siloed roles to create better performing and more reliable products.
After several years of DevOps practice, the business felt a strong interest in integrating with the various practices. In this evolution, BizDevOps was created. Just as DevOps bypasses the silos that separate development from operations, BizDevOps bypasses the silos that separate DevOps from the rest of the business. To accomplish BizDevOps, collaboration between the business team and product managers, developers and operators to prioritise sprints or backlogs must be fostered. It is important to encourage this collaboration throughout the release cycle.
And finally, another discipline that we will detail in a future article is Remediation, codenamed NoOps. This practice consists of the complete automation of IT management, often based on Artificial Intelligence. It is about managing infrastructure and application deployment while eliminating the need for human intervention.
