Home › Blog › Von Arno Lais veröffentlichte Artikel

SupplyOn Blog

Arno Lais

Vice President IT-Operations

Als Verantwortlicher für den Betrieb und weltweiten Kundensupport der SupplyOn Platform, liegt mein Fokus auf der 100%tigen Systemverfügbarkeit, der Integration moderner Technologien sowie der nachhaltigen Verbesserung von Prozessen und Servicequalität. Durch jahrzehntelanger Erfahrung in der IT-Branche bin ich überzeugt, dass Erfolg in der IT nicht nur von Technologie abhängt, sondern ebenso von klaren Strukturen, guter Kommunikation und einem starken Teamgeist.

Beiträge von Arno Lais:

Beiträge dieses Autors in anderen Sprachen:

Technology

Observability: A step towards 100% SLA availability

In today's digital world, high service level agreements (SLAs) are a key competitive advantage. When using internet platforms, customers expect them to operate without interruption. For many businesses, even short downtimes can have serious consequences. Of course, no technology is immune to failure. So 100% availability is a very ambitious goal. A key component in achieving this goal is the implementation of observability as a core principle of system monitoring and optimisation.What is the key success factor of observability for 100% availability?Unlike traditional monitoring systems, monitoring methods are usually "reactive". In the worst case, this means that a failure is only detected when it has already occurred. A comprehensive observability strategy detects potential problems before they cause failures. Teams quickly identify and resolve anomalies in metrics like latency or memory usage.If a problem occurs, observability helps to quickly identify the possible cause. Traces provide precise information about where delays or errors occur, while logs provide the necessary details to analyze the cause. By continuously analyzing system metrics and logs, bottlenecks and vulnerabilities are detected before they affect availability. This means that optimization is proactive. In practice, logging an application is a major challenge. Logging data must be defined and implemented by the application developers together with the operating units, among others. Here, the DevOps concept plays an important role again, which must be implemented “hand in hand” in practice. Observability should therefore be an integral part of the DevOps workflow.Modern IT systems often use microservices that run in both the cloud and on-premises data centers, with different teams managing them based on the application. Observability creates transparency and provides a better understanding of dependencies and potential vulnerabilities in these complex architectures, promoting a culture of continuous optimization.Challenges in implementing observabilityThe most common problems include selecting the right tools for analyzing mass data. Key technologies include APM (application performance monitoring) tools with AI functions developed specifically for observability. Due to the high volumes of data generated by monitoring, it is necessary to use high-performance databases and event bridges to bundle events and provide detailed control over the delivery of messages to different recipients.ConclusionObservability is an indispensable tool for companies striving for the highest availability. It is not a one-off task, but an ongoing process. It enables organisations not only to react to problems when it is too late, but also to counteract them proactively. Teams can overcome potential organizational resistance by clearly communicating the benefits and added value. The investment in observability pays off: it increases the stability and performance of systems, bringing you a decisive step closer to the goal of 100% SLA availability.
Arno Lais · 27. Januar 2025 - Lesedauer < 3 Min.
Observability: A step towards 100% SLA availability