With the rapid development of cloud native technologies, application architectures, infrastructure, and software life cycles have changed dramatically. The application architecture transforms from single application to microservice and becomes a loosely-coupled system. Kubernetes becomes the default platform for running containers, with IaaS, PaaS, and CaaS platforms underlying the Kubernetes platform. Services are continuously deployed through the DevOps pipeline, and service changes are less costly, less risky, and more frequent.
While cloud native brings a lot of convenience for development and operation, it also makes monitoring more complicated: the scale and dynamic nature of microservices make data collection costs significantly higher. It is difficult to quickly locate the root cause of large-scale service problems. For visualization and correlation analysis, traditional operation and maintenance monitoring lacks corresponding means.
Traditional monitoring is more based on the external perspective to monitor the whole system, pay attention to part of the system behavior, and plan the failure model of the whole system from the perspective of operation and maintenance. With the popularity of cloud native, the concept has been expanded from monitoring to observation, and attempts have been made to observe the internal operating status of the system from the idea of white-box system. Meanwhile, multiple observation means have been combined for in-depth analysis to understand the root cause of the operating status of the whole system.
More and more teams are realizing that they should proactively plan relevant observable capabilities from the beginning of technical architecture design and construction, including: support rapid business iteration, support complex call topologies, support user experience optimization, and efficient operation and maintenance collaboration. In the face of challenges and demands, how can enterprises better practice cloud native while building their own observability?
In order to solve the above problems, Ali Cloud native application platform has created an observable month in December, and launched “Ali Cloud Observable Series open classes” for 4 periods. One product and technical expert will be invited every week to comprehensively interpret the practices and solutions of observable scenarios from multiple dimensions to help enterprises comprehensively strengthen their own observable capabilities.
Click here to sign up for the first live broadcast on December 9th – “All in One: How to Build an End-to-end Observable System”!