Shulou(Shulou.com)05/31 Report--

This article introduces the example analysis of SkyWalking distributed link tracking, the content is very detailed, interested friends can refer to, hope to be helpful to you.

SkyWalking is an open source APM system, including monitoring, tracking, and diagnosing functions for distributed systems in the Cloud Native architecture. The core functions are as follows:

-Analysis of service, service instance and endpoint metrics

-Root cause analysis, analyzing code at run time

-Analysis of service topology diagram

-Service, service instance and endpoint dependency analysis

-slow services and endpoints detected

-performance optimization

-distributed tracking and context propagation

-Database access metrics. Detect slow database access statements (including * * SQL** statements)

-call the police

SkyWalking is currently a top-level Apache project, as such an excellent open source project, its architectural design concept will certainly have a lot to learn from.

Ecological Methodology of message Middleware selection

Middleware is software that provides general services and functions for applications. Data management, application services, messaging, authentication, and API** management are usually done through middleware. Middleware can help developers build applications more effectively. It is like a link between applications, data and users.

For enterprises with cloudy and containerized environments, middleware can help you develop and run applications on a large scale and cost-effectively.

Middleware is software between the operating system and the application running on it. In essence, middleware acts as a hidden transformation layer, realizing the communication and data management of distributed applications. It is sometimes called a pipe because it connects two applications together so that data and databases can be easily passed between pipes. Through middleware, users can perform many requests, such as submitting a form on a * * Web** browser, or allowing a * * Web** server to return a dynamic web page based on the user's profile.

Common examples of middleware include database middleware, application server middleware, message-oriented middleware, * * Web** middleware, and transaction monitor. Each program typically provides messaging services that allow different applications to communicate using messaging frameworks such as simple object access Protocol (* * SOAP**), * * Web** services, declarative state transition (* * REST**), and * * JavaScript** object representation (* * JSON**). Although all middleware perform communication functions, the type chosen by the company will depend on the service to be used and the type of information to be communicated. This can include secure authentication, transaction management, message queuing, application servers, * * Web** servers, and directories. Middleware can also be used for distributed processing of real-time operations rather than sending data back and forth.

What I want to focus on here is message middleware, which can be said to be often contacted and used in our daily development, which belongs to the conventional technology stack.

Message middleware, then I will not compare one by one here, because there are too many articles compared on the Internet, and there are only two mainstream ones that are used very frequently at this stage, * * Kafka** and * * RocketMQ**.

Briefly explain the scene, more incisive. * * Kafka** is suitable for high throughput, and then can tolerate some message loss, such as log collection scenarios. * * RocketMQ** is used in business scenarios with high reliability and certain throughput, such as financial payment scenarios.

The methodology of middleware selection is actually very simple, which is considered in terms of the degree of matching between business scenarios and middleware functions, as well as the cost of maintaining such a system. From a long-term point of view is expansion and cost, from a short-term point of view is also income and cost.

We usually judge a message middleware from a technical point of view, such as high availability, high performance, high concurrency and high throughput, but in fact, any middleware that can be opened up is not bad. The key point is how we use it. And be able to understand the principle, so that it can really give full play to its value. I have seen many teams often complain when using open source message middleware, saying it is bad. It is very difficult to use, but you have to think about it. If you write it yourself, the message middleware written in half a year may not be as good as others', or there may be bug**, everywhere because you do not have more business scenarios to verify. Open source message middleware has at least been implanted into more and more extensive business scenarios by people in the community.

This is the end of the example analysis on SkyWalking distributed link tracking. I hope the above content can be helpful to you and learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.