What would you choose between architecture optimization and business iteration?

Introduction

For every software system , We can all reflect its value through business and architecture .

Especially software developers , You should ensure that the actual value of your system in both dimensions can be maintained at a high level for a long time .

But it's a pity , They may focus more on one dimension , And ignore the other ;

If you focus on the wrong dimensions , This leads to a decline in the value of the system to 0, Very regrettable .

1、 Business and technology issues

In our daily work , Business iteration support and system architecture technology optimization are like fish and bear's paw , You can't take care of it at the same time .

case1：

When we find that the system performance is a little poor , The assessment needs to be optimized , Reduce system interface noise , At the same time, improve the user experience ...

But at this time, the product students ran over and said , Recently, there is a business revision requirement / New business functions need urgent online support .....

Q： What would you do ？

-- If it weren't for technical optimization , Let's wait until this requirement is completed ...

case2：

The technical planning we have to do at the beginning of each year , Do you always feel that the plan can't keep up with the changes ...

Every middle of the year 、 At the end of the year , Reality is always very different from planning , Either the planning is not done or a low-cost solution is simply implemented , Or something new happened in the middle of the way that disrupted the original plan .

Q： Do it again , How would you avoid ？

-- planning vs change

The above situation should be very common in work , What principles should we follow when facing similar problems ？ Is there any effective solution ？

2、 Which is more important, business or technology ？

The former president of the United States, Eisenhower, put forward a classic “ emergency / The importance matrix ”.

I have two problems ： Urgent and important , And urgent problems never matter , Important problems are never urgent .

Q： Which dimension is more important ？

In the eyes of R & D students , The importance level is clear at a glance ：

The important thing ： Architecture design optimization , Make the system have enough “ play ” sex

Something urgent ： Business iteration support , Let the system support the sustainable development of the business

for example ： Daily business iteration support is always urgent and excellent , But from the perspective of rational construction of Architecture Design , There doesn't seem to be so many connections ; And system technology optimization seems to be very important , But often there is no chance to arrange the last issue .

About which one is more important , Of course, different people have different opinions

If for this question , It's up to the business department to answer , That's business is more important - The system supports business iteration , It is very important to ensure the normal development of business .

3、 What more attention should be paid to R & D ？

We can try to sort these four cases ：

1） Important and urgent

2） Important but not urgent

3） Not important but urgent

4） It's not important and urgent

System architecture design optimization ： important （ Occupy the first place 1、2 position ）

Business iteration support ： emergency （ Occupy the first place 1、3 position ）.

But in our daily work , The common mistake that business departments and R & D departments often make is to raise the third priority to the third . The important issues of system technology optimization are excluded by business iteration .

Our R & D personnel often complain about , There's no time for technology optimization , I tease myself for ：“ Moving bricks again ...”、“ Working overtime to write BUG 了 ...”

It seems that we have forgotten ： The business department is just focused on the business , Evaluation and optimization of system architecture , It's the responsibility of R & D personnel ！

How to balance the work of the two , It's the way for R & D personnel to be promoted and cultivated .

Don't ignore the value of system architecture , If one day the system is so hard to maintain that it can only be overturned and restarted , It can be said that the system technology optimization can not keep up with the rapid iteration of the business , At the same time, it shows that the R & D students are not qualified enough in their own work .

4、 Practical solutions for reference

There are many difficulties in balancing technical architecture optimization and business iteration support . Is there a good way to balance ？

I collate some experience from my own work practice , Mainly “ total - branch - total ” Principles , For your trial reference ：

4.1、【 total 】 Project approval , Evaluate the target return

Do things with value , In particular, technical optimization projects , Be sure to understand what the payoff point is , At the same time, we should make clear the cost performance of input and output .

It is suggested to spend more time at this stage , More research and analysis , It is suggested that the goal should be as quantifiable as possible , Achievable .

【 Project goal setting - Example 】

project ：XX System performance optimization

The goal is ： System service level sound <= 200ms（90 Quantile value <= 500ms）

4.2、【 branch 】 Split and refine , Evaluate the scope of optimization transformation

After setting a goal , The target needs to be disassembled , And evaluate the percentage and priority of each action .

Function point splitting should start with the end , Make sure you don't deviate from the goal .

Specific methods can be referred to ：

1） Split according to process logic

2） Split by work module

3） Split by stream

It is possible that the previous assessment of each point may not be accurate , Allow for later adjustments , But there should be no big adjustment , Otherwise, it needs to be reconsidered .

【 Project breakdown - Example 】

Concurrent 、 Service dependency optimization 、cache、 Table structure split 、 Table index optimization 、 Logic optimization （ Loop set loop ）、Redis Big Key Optimize 、 Resource isolation wait

4.3、【 total 】 Categorical convergence , Plan feasible project milestones

Functional assessments are classified by the same type , Split the sub optimization items into daily business iterations , Or interpose into the gap of each requirement development , Small step run , The project was steadily promoted .

【 Categorical convergence - Example 】

4.4、【 guarantee 】 Report and follow up , Project iteration progress

A weekly project reporting system can be established , Call relevant students to follow up on a regular basis , Project progress & The problem continues Review, Ensure the overall progress of the project .

The main focus of project progress report ：

summary

Many people refer to the chaotic relationship between business and technology as “ Love that kill ”, In fact, technology and business are not antagonistic , It's complementary .

Technology as the theoretical basis , Business as a practice to test landing . The technology we learned , Business is our landing point , Business success can also reflect the technical value .

This article is from WeChat official account. - The road to the advanced architecture （jiagou_jingjin）.
If there is any infringement , Please contact the [email protected] Delete .
Participation of this paper “OSC Source creation plan ”, You are welcome to join us , share .