In multi-cloud, multi-datacenter environments, IT needs new methods for monitoring and troubleshooting applications. Observability tools can provide that.
In the early 1960s, Rudof E. Kalman, a Hungarian-American engineer, introduced the concept of system observability, defined as “a measure of the degree to which the internal states of a system are measured. system can be inferred from its knowledge. external output. ”
In there, visibility networks and systems are facilitated by “debug” utility programs that are bundled with the software or developed in-house, and subsequently by network management software.
Over time, more predictive analytics and automation have been incorporated into engineering workflows, with network and system monitoring and warning of flagging issues (such as failed drives) long before they become problems so IT can resolve them without service disruption.
All of these are mature and successful approaches that have worked for many years in more monolithic platforms and stacks of traditional computing. However, now with the shift to hybrid computing, where applications cross multiple clouds and even data centers, it is not possible to know every step of what is going on in an application workflow. use. Why? Because the system and network monitoring utilities used today are not designed to go through different computer platforms and track every action. What they were (and are) good at is creating logs of what’s going on in a particular system or network.
UNDERSTAND: Ethical policy on artificial intelligence (TechRepublic Premium)
“The move from monolithic applications to microservices can be overwhelming,” said Charlotte Dunlap, principal analyst for application platforms, technologies and enterprise services at GlobalData PLC. “A big challenge is that the task of monitoring and optimizing applications running in a container environment like Kubernetes is almost impossible. As a result, DevOps effort participants are forced to re-examine how they implement observability earlier in the application lifecycle to improve insights in the future. basic infrastructure. ”
Corporate IT departments need to address this if they want to achieve comprehensive visibility (and troubleshooting capabilities) in multi-cloud and data center environments. This is where software visibility comes in.
Like network and system monitoring software, visibility software will provide you with logs allowing you to troubleshoot problems in any network and silos. However, observability software goes even further by taking advantage of full tracking of all application activities and components across a combined network of multiple data centers and clouds that a single application can access. may be in use. Along the way, the software observes the performance of each application component.
UNDERSTAND: Metaverse Scam Table: Everything You Need to Know (Free PDF) (TechRepublic)
The end result is that IT gets logs and performance metrics for how each application component on the combined chain of cloud and data center IT infrastructure is performing — and it also gets logs. detailed operations for each component.
This restores end-to-end visibility of application performance in hybrid cloud and data center infrastructures, and enables IT system programmers and network administrators to address and faster application performance tuning. An artificial intelligence (AI) layer can also be added to the observability which speeds up problem resolution time because it removes the “noise” that the system and network alerts would flag, but that’s really not the case. not a problem for problem solving.
It is now up to IT to adopt these new tools that can re-establish 360-degree visibility for applications in multi-cloud and data center environments. Statista estimates that by 2024, the global observability market will be at 19.8 billion dollarsso it looks like companies are getting the message across.
