Let me give you an example before answering this question: Does a programmer need to learn mathematics?
Some developers say why learn math. Aren’t the things we have now enough? Writing code doesn't require math.
I want to say that the algorithm of people who say this must not be good!
Taking the recommendation system as an example, collaborative filtering algorithm, dimensionality reduction algorithm and interactive evaluation-based recommendation algorithm
For dimensionality reduction algorithms, you have to first establish a coordinate system and then assume equation modeling. Find the mapping variance
For example, the collaborative filtering algorithm assumes that the books that user 1 likes are A, B, and C. The books that user 2 likes are A, C, E, F. The books that user 3 likes are B and D
The problem we need to solve is: decide whether the book F should be recommended to 1. Then you first need to know what the types of A, B, C, D, E, F are and then model them (discrete, mathematical modeling)
Then when a developer writes back-end or front-end (mobile app), he only writes the interface and some login and registration functions, as well as content publishing. These do not require mathematics. So if you are a developer of a singing bar (app) and you want to recommend songs based on the songs sung by users, will you use mathematics?
Finally, back to this question, you don’t need to consider clustering or distribution for a small number of temporary users, but if your user volume increases sharply. And you don’t have enough funds. what would you do?
Docker has a wider range of uses in the actual production environment. It can use very few resources to build a cluster. If it is only used in the development environment, it is really a waste of resources.
It is still necessary. After all, docker is still different from a physical machine. It is impossible to just operate and maintain it after development. Docker is very useful for environment construction, and it is necessary to learn it.
At least it is a way of solving problems, and it is not bad even as a knowledge reserve.
Why use docker? Docker is very powerful in deployment. If you are doing development, you may still be lacking something. For example, use strace to track the process in the container. Although you can turn on the privileged mode to use it, there are still many pitfalls in the privileged mode that have not been solved.
Personally established laravel environment project: https://github.com/Jamlee/docker-laravel
Compared with VM, docker has obvious advantages in its lightweight, configuration complexity and resource utilization. As docker technology continues to mature, more and more companies are beginning to consider using docker to improve their IT systems.
Now there are also a number of docker-related startups emerging in China. Among them is a container management project - csphere https://csphere.cn/ You can pay attention to it. Get a feel for the current application scenarios of docker, etc.
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
Let me give you an example before answering this question: Does a programmer need to learn mathematics?
Some developers say why learn math. Aren’t the things we have now enough? Writing code doesn't require math.
I want to say that the algorithm of people who say this must not be good!
Taking the recommendation system as an example, collaborative filtering algorithm, dimensionality reduction algorithm and interactive evaluation-based recommendation algorithm
For dimensionality reduction algorithms, you have to first establish a coordinate system and then assume equation modeling. Find the mapping variance
For example, the collaborative filtering algorithm assumes that the books that user 1 likes are A, B, and C. The books that user 2 likes are A, C, E, F. The books that user 3 likes are B and D
The problem we need to solve is: decide whether the book F should be recommended to 1. Then you first need to know what the types of A, B, C, D, E, F are and then model them (discrete, mathematical modeling)
Then when a developer writes back-end or front-end (mobile app), he only writes the interface and some login and registration functions, as well as content publishing. These do not require mathematics. So if you are a developer of a singing bar (app) and you want to recommend songs based on the songs sung by users, will you use mathematics?
Finally, back to this question, you don’t need to consider clustering or distribution for a small number of temporary users, but if your user volume increases sharply. And you don’t have enough funds. what would you do?
You have to stand tall to see far!
If you consider the future, it is still necessary. After all, the more you understand, the more choices you will have in the future.
DevOps practices.
Docker is largely an operation and maintenance management tool for developers.
Of course, if you don’t need to publish a complete project and don’t have your own Web App product, you really don’t need Docker.
Docker has a wider range of uses in the actual production environment. It can use very few resources to build a cluster. If it is only used in the development environment, it is really a waste of resources.
I only use vagrant for the development environment. It is very convenient. Docker is not necessary
It is still necessary. After all, docker is still different from a physical machine. It is impossible to just operate and maintain it after development. Docker is very useful for environment construction, and it is necessary to learn it.
At least it is a way of solving problems, and it is not bad even as a knowledge reserve.
Why use docker? Docker is very powerful in deployment. If you are doing development, you may still be lacking something. For example, use strace to track the process in the container. Although you can turn on the privileged mode to use it, there are still many pitfalls in the privileged mode that have not been solved.
Personally established laravel environment project: https://github.com/Jamlee/docker-laravel
Compared with VM, docker has obvious advantages in its lightweight, configuration complexity and resource utilization. As docker technology continues to mature, more and more companies are beginning to consider using docker to improve their IT systems.
Now there are also a number of docker-related startups emerging in China. Among them is a container management project - csphere https://csphere.cn/ You can pay attention to it. Get a feel for the current application scenarios of docker, etc.