I have several friends who are engaged in PHP development, and I have also come into contact with many PHP engineers. They often doubt their future technical growth and development. I often give them some suggestions, hoping that they can make breakthroughs. themselves, have better development.
First of all, let me be clear about the PHP engineering questions I am referring to. I mean that after graduation, I mainly used PHP to develop WEB systems and have never worked in other languages. I have about 3 to 4 years of work experience. I am basically comfortable in developing ordinary WEB systems (million-level access, less than 1,000-level data, or the business logic is not particularly complex) without any problems. But they will have characteristics like this:
These PHPers lack ideas when they encounter projects that require high performance, high concurrency processing, and large amounts of data, or the business logic is relatively complex (the system needs to solve business problems in multiple fields). They are unable to analyze the essence of the problem and have poor technical judgment. They can quickly find temporary solutions to the problem, but often the system and themselves collapse step by step in the constant temporary solutions. So how do you improve yourself? How can I challenge a more difficult system?
Where are the higher challenges? Based on my own experience, I list some specific challenges to give everyone a perceptual understanding.
What are the challenges of high-performance systems?
What are the challenges of high-complexity systems?
When you can answer the questions I listed in the affirmative, I think it is technically possible for you to become an architect. If you can't answer it yet, you need to strengthen in the following directions.
If you can’t answer it yet, you need to strengthen in the following directions:
"How do you learn so many things? How long does it take to learn?" If you work hard and have a good plan, it will probably take 1 to 2 years.
How to study effectively is a big problem. I have some practice but it is very scattered and difficult to summarize. Before going to bed last night, I suddenly thought of the core of RUP, which is "architecture-centered, use case-driven, iterative development." Borrowing this idea, effective learning methods can be expressed like this: centered on principles, models or mechanisms, Task-driven, iterative learning.
It’s a bit abstract, let’s give an example to illustrate how to learn. Purpose: Learn how to improve processing performance.
Iterable-driven task: Find the location by IP.
This is a common task for WEB applications. The IP database has about 100,000 rows of records.
First iteration: Implement functionality without considering performance (implemented through PHP). Because the region cannot be searched directly through KEY (IP), simple methods such as directly placing it in the data or through an associative array will not work. The idea is to sort the data first and then search.
Learning objectives: Sorting algorithms, search algorithms.
PHPer generally has a poor foundation in data structures and algorithms. He usually does not have tasks in this area and does not study himself, so he lacks knowledge in this area. However, the problems solved by programming will ultimately come down to the data structure and the algorithms that operate on this data structure. If the data structure algorithm is always in your mind, you will be able to clearly understand its inner structure when you encounter a problem, and the solution will naturally emerge.
Second iteration: Optimize data loading and sorting. If you do the first step, it is basically not available because the data needs to be loaded and sorted every time, which is too time-consuming. The solution is to load and sort the data once and put it in a place where each PHP process can access it.
Put it in memcache This is an easy question for everyone. In fact, putting it in shared memory (supported by EA and other accelerators) is a faster way, because memcache also has more network operations. Is the data put into shared memory as a whole or in blocks? How to test performance? How to analyze bottlenecks (xdebug)? Driven by these issues, you will learn.
Learning objectives: Methods to detect, locate, and optimize PHP performance; the impact of PHP implementation structure on performance.
Third iteration: Writing extensions to PHP. The performance still cannot improve, and I have to enter the world of C/C++. But from now on, you will not only be a PHPer but also an all-round engineer on the server side. Of course, this will be a huge challenge for students who have never done C/C++. I can't simply explain how to learn C/C++ here. You can refer to "PHP Programmers Learn C++"
Learning goals: Learning C/C++, writing PHP extensions
How to determine the mechanisms and principles that need to be learned? How to find the driving learning tasks? I don’t have any idea about what I need to learn. How can I answer the above two questions?
If I need to learn Javascript, I have some perceptual knowledge of HTML and CSS. First of all, I learned that JS is a dynamic language in the WEB field, which mainly solves the dynamic interaction of web pages. Then the key points I want to learn are as follows:
If you are completely self-study, find the key points that need to be learned (mechanism, model, principle). Setting learning tasks is indeed not that easy to grasp. If you find an experienced person to guide you or add a learning team, the speed of learning will indeed be greatly improved.