Shulou(Shulou.com)06/02 Report--

This article is excerpted from the book "Crazy handouts for Xiaoqiang Software testing"

This topic is a bit heavy, because if it is not well expressed, it will certainly be scolded by some unscrupulous people, but I decided to write this chapter after thinking it over. Because I have been asked this question by too many friends, about statistics, basically every two days will be asked, sometimes I will be asked N times a day, I will answer the hematemesis, for this in the "get kicked talk Show" (http://www.lizhi.fm/200893/) to do a special program, we can see the necessity of this topic, but also hope to help friends with choice entanglement.

Below I try to objectively talk about my own learning experience, perhaps this experience is not the best, or even wrong, but can give you some reference, less detours, I think it is valuable.

First of all, let's talk about several challenges that learning performance testing needs to face. You can combine your own actual situation to see if you are suitable to continue learning.

First, the huge knowledge system, this is the first challenge we face. Performance testing is a complex and patient task, and we need to "peel off the cocoon" in the complex system and analyze layer by layer to determine the performance problem. This process will involve middleware, Web server, cache, database, code and other knowledge, so it is difficult to continue without a more complete knowledge system. Although it is a challenge, in my opinion, it is the best way to get started for most rookie friends, because it can help us quickly establish a more perfect knowledge system, and it will do us no harm. I don't know if you have ever encountered such a scene, being pointed at the nose and saying: you can't even write a SQL statement, and you don't even know what the middleware is. Although such "humiliation" makes us unhappy, it also bluntly points out the lack of the overall knowledge system of many test engineers. Only by making up for their own shortcomings can we have the strength and strength to strive for better things.

Second, strong analytical skills, which is the second challenge we face. Just like the cartoon "Conan", solving a case at a complex crime scene requires constant inference and argumentation. in this process, it may overturn what has been determined before, or there may be no progress for several days, but this is also its charm. It can be said to be painful and happy.

After I have come into contact with many students, I find that a common problem is the poor ability of logical analysis. In the process of analysis, it is often a little bit from east to west, there is no logic at all, it is random guessing, and it is often easy to fall into details. Once you fall into it, it leads to stagnation, which is why many people find performance testing difficult. In my opinion, the analysis process of performance testing is like peeling onions, you need to peel layer by layer to see the problem, this process requires you to have strong logical analysis ability, but also macroscopical. Only stand at a certain height to look at the problem can suddenly be enlightened, otherwise you will fall into a dead end. Once this thinking ability is cultivated, it will get twice the result with half the effort, and the efficiency will be improved when learning other technologies, so you have to pay for everything in order to get a harvest.

Second, let's talk about several challenges that we need to face in learning automated testing.

First, the coding ability, this is beyond the limit. Speaking of which, some friends may ask, doesn't performance testing require coding ability? The answer is yes, but the threshold is relatively low compared to automated testing. In fact, coding ability is a necessary skill for a good test engineer.

How to improve their coding ability is also a question that many friends have consulted me, and there are really no shortcuts. I think we should practice more and sum up more. The practice I'm talking about is really doing it instead of watching it. In fact, most of the students I have brought have a problem, that is, when I speak in class, I feel very simple and disagree, but when I practice after class, I have all kinds of problems, and very simple knowledge points can be done for a day. So you must practice more, sum up every mistake you have made, and don't let yourself fall twice in the same place. I would like to say again: "there is no inconspicuous brick, no invisible frame, how can a beautiful building stand."

Second, the ability of logical thinking. Can you do automated testing after you have the coding ability? Obviously not, because automated testing ultimately hopes to build a framework or platform, which is a big project, must have strong logical thinking ability and design ability. For example, just because you can weld doesn't mean you can design cars. So the real difficulty of automated testing lies in the design idea, and it will be difficult for friends who have no experience to do it, which is why I personally suggest that we can learn performance testing first. the reason why you learn automated testing after you develop your ability and thinking.

Having said so much, I think members should already have the answer in mind. I would like to state once again that these are only my personal views. They are not necessarily right. They are for reference only. Do not spray them if you do not like them.

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.