Upstream and downstream in software development

  After listening to Tencent rhinoceros bird Wang Jia's explanation of open source , Upstream of the technical terms （upstream） And downstream （downstream） incomprehension , Read this article blog I understand that the records are hereby reproduced .

Look at the following blog I understand ： In nature, water flows from high to low , The kinetic energy of the water at the bottom is higher than that at the top , Abstract into value , So value （ Kinetic energy ） The relatively high is downstream , On the contrary, it is upstream .

What is Upstream and Downstream in Software Development Chinese translation of .

In the last period of time , I began to come into contact with “ The upstream ” and “ The downstream ” These two words , Often confused . Every time I have to check its meaning , So I decided to write them as blog To help understand .

Upstream and downstream in the production process

---

Let's start with a simple production process , Although it has nothing to do with software development , But we can define the upstream and downstream of software development on this basis .（ Activity diagrams , A solid indicates the beginning , The middle point means the end ）

In the example above , We have three steps :

1. Collect parts
2. Assemble parts
3. Draw the assembly

The above production process is very similar to the river , So it's easy to understand   As the process moves from one step to the next , Gradually " The downstream " Move

We can infer the following rules :

1. Rely on the rules : Each current project depends on all upstream projects
2. Value added rules : In the process of moving downstream , Each step adds more value to the product

Now? , Let's try to apply these rules to different software development environments .

Upstream and downstream of software dependencies

---

Most software components depend on other components . So what are upstream dependence and downstream dependence ？

As shown in the figure :

Components C Component dependent B, And components B And depends on components A.

application Rely on the rules , We can definitely say components A It's a component B Upstream , And components B It's a component C Upstream （ Even if the arrow points in the other direction ）.

Apply here Value rules A bit abstract , But we can say components C Have the greatest value , Because it “ Import ” Components B and A All the characteristics of , And add your own value to these features , Make it a downstream component .

Upstream and downstream open source projects

---

The other is often used “ The upstream ” and “ The downstream ” The environment of these two words is open source development . It is actually very similar to the component dependencies discussed above

Consider the project A and B, among A It's the original project ,B yes A The branch of ：

This is a fairly common development style in open source projects ： We create a branch of the project , Fix the error or add functionality to this branch , Then submit the patch to the original project .

under these circumstances , Rely on the rules Make the project A Become an upstream project , Because it can be in no project B To survive well , But if there is no project A（ Original project ）, project B（ Bifurcation ） It won't even exist .

Value rules It also applies here ： project B Added new features or bug fixes , It is the original project A Added value .

therefore , Every time we contribute a patch to an open source project , We can all say We have sent a patch to the upstream .

Upstream and downstream in the microservice architecture

---

By microservices （ Or just old-fashioned distributed services ） In a system composed of , There are also discussions about upstream and downstream Services .

Rely on the rules and Value rules It also applies to this situation

service B It is an upstream service , Because the service A Depend on it . service A It is a downstream service , Because it adds Services B The value of .

Please note that , In this case, the upstream and downstream definitions of “ flow ” Not through service A Data flow into the system , It is Data flow system to user oriented service .

The service is far away from the user （ Or any other end consumer ） Closer , The farther it is from downstream .

Conclusion

In the use of “ The upstream ” and “ The downstream ” In any context of the concept , We can apply two simple rules to find out which project is upstream or downstream of another project .

If one project adds value to another project or relies on it in any other way , It must be downstream .