Question PreviewThrough interviews and manuscript requests, veterans in the field of operation and maintenance are invited to provide profound insights and collide together, with a view to forming some advanced consensus and promoting The industry has to move forward in a better way.
In this issue, we invite Zou Yi, the Operation and Maintenance Director of Tuyou Games. Mr. Zou often jokingly calls himself the Operation and Maintenance representative of the world's top 5 million companies. It can be seen that in his heart, he feels that the Operation and Maintenance of small and medium-sized companies is very important. The construction ideas are different from those of large enterprises. Today we asked Mr. Zou to share his approach to the integration of research and operation for small and medium-sized companies with a few questions.
This is the 6th issue of the down-to-earth and high-level "Operation and Maintenance Forum" , let’s start!
Q: Tuyou is a game company, what do you think are the unique features of game operation and maintenance? What are the biggest operational challenges you face? How did you solve these challenges?
The overall game operation and maintenance architecture is relatively simple compared to traditional Internet business, but the single-machine reliability requirements are relatively high. The daily operation and maintenance work involves more transactional work, such as launching Service combination and so on. The biggest operation and maintenance challenge we face is actually not at the technical level, but more at the value recognition level. How to make our business departments recognize our value. I believe this challenge is also a unanimous challenge for colleagues in the entire operation and maintenance track. To win the recognition of the business department and enhance the value of the operation and maintenance team, I can sum up the practice of myself and my team in one sentence:Do a good job in down-to-earth service, with the business department/user as the center.
Q: What are the talent skills for game operation and maintenance? If you want to develop in the direction of game operation and maintenance, do you have any suggestions for career path planning?
The talent skills of game operation and maintenance are not much different from those in the traditional Internet industry. For the operation and maintenance track,low awareness and lack of systematic growth environment , is a more realistic problem faced by our small and medium-sized factories in operation and maintenance. We deal with the bottom layer of the machine all the year round, and rarely think about it seriously. The development in the next 10 or 15 years will be more about chasing hot spots and chasing changes. , rarely think about the accumulation of unchanging content, and how to use this content to make time for friends to form their own competitiveness. I personally suggest that operation and maintenance students in small and medium-sized factories should focus on both theoretical methodology learning and skill improvement, use theory to guide practice, and improve their understanding of theory through practice. I would also like to make some suggestions regarding learning theories and methods:
Question: The operation and maintenance team of a medium-sized company is usually not very large. How do you organize the limited manpower? Do you have any experience to share? Everyone?
Limited resources are often easy to stimulate innovation. The team size does not need to be large, but it must remain lean and agile. In other words, your team must be able to fight and cope with uncertainty. The ability must be strong. To achieve this effect, I personally summarized our organizational capacity building practice in the past five years:
Regarding the practice of agile organizations, you can read my sharing: https://tuyoo.feishu.cn/docs/doccnFlAD2m7WnSpcLYxFJRImZb
Q: Are you You may encounter problems where your ideas are slow or difficult to implement because of the poor talent level of your team. How do you solve this problem?
This will definitely happen. Our solution is:
About my team transformation practice sharing: https://tuyoo.feishu.cn/docx/doxcnGMuijglK6NdENYC2vD7KKh
Q: You said that you particularly agree with "Yun" According to the viewpoint in the article "The Future of Dimension is Platform Engineering", your team is also a full-featured production and research-type organization. I would like to ask you to introduce: For business research and development, compared with directly using platform products provided by cloud vendors, your team leads What is the delta gain coming from?
Before answering this question, I still want to elaborate on our understanding of wheel manufacturing and external procurement services:
In fact, we do external procurement or self-research, quite a bit An open mind, and a fairly simple judgment, is to look at the input-output ratio of ROI. If it is standardized and the investment is huge, if we cannot solve it ourselves, we must try to use the services or products of external third parties to help us solve it. Question, We are more concerned about how to serve our business departments well, and the results and quality of the services we provide. We are not too concerned about whether this capability is our own or the service capability of a third party. As long as it can help us To improve service quality and efficiency, we are very open-minded to absorb and integrate.
Let’s answer the question about the gain of this production and research team for us. Each company has its own characteristics or customized scene requirements. These things cannot be fully covered by foreign products, so such an end-to-end team The end team actually gives the entire team the ability to solve some non-standard problems. This ability is actually very critical and determines the value realization of the team to a large extent.
In addition, let’s talk about our understanding of The future of operation and maintenance is platform engineering. My understanding of platform engineering has two key elements:
Our team’s transformation exploration is mainly based on these two elements. However, the theoretical level is not enough and we have not clearly proposed the concept of platform engineering. A big pain point in our game operation and maintenance is that there are many trivial matters, such as CDN upload and release, game configuration updates, routine start and stop of servers, which are all daily tasks of game operation and maintenance. They are indispensable, but they are all transactional. , the value is very low. Maybe in our common sense of game operation and maintenance, we will think of making some automated tools to improve the human efficiency of operation and maintenance, and change the operation and maintenance from the state of human flesh or writing scripts to the state of WEBOPS. This feeling The leverage ratio is still too low, and operation and maintenance is not released. Therefore, in the process of solving these problems, my original understanding of the platform engineering concept was born. Currently, 50% of the daily routine work of our game operation and maintenance is performed by the project team. Self-service, through the tools we provide, is something we found to be highly cognitively consistent after being exposed to the idea of platform engineering. Therefore, the future of operation and maintenance is platform engineering. I believe that students who have tasted the benefits of self-service and experienced the hardships of human-powered operation and maintenance will have a deep sense of identification.
Question: You often say that cost savings require hard work and hard work, saving a lot of costs. The company issued a certificate of award, indicating that this FinOps project is most likely to be self-inflicted and in the cloud. In the construction of Infra on the cloud and under the cloud, your team has brought huge cost savings to the company and received material rewards from the company. Can you share your relevant experience?
Regarding FINOPS, I usually have some exchanges and discussions with some experts and teachers in the industry. Combined with our team’s own practice, I personally feel that the implementation of FINOPS is difficult, and the difficulty lies in changing the boss. At present, the industry is still in the stage of technical realization or concept collision, and it is still in the stage of comparing who is more professional and more standardized. I personally feel that FINOPS that cannot affect the boss’s cognition are basically worthless or of extremely low value. There is no difference between doing it and not doing it. I don’t want to comment too much on the field of FINOPS. We have narrowed it down to cost optimization. In our team, I have not set OKRs for cost optimization. We have always used the concept of lean to guide our work. Lean has a core concept. Everything that does not produce value is waste. We must continue to eliminate waste. In this way, there is no need to engage in dynamic cost optimization during the work process. Many cost optimizations that have saved hundreds of millions may be the right ones in the eyes of the boss. In the past, there was too much waste, but now we just eliminate waste, which will naturally not be recognized as valuable.
I have personally summarized a few points during the cost optimization practice:
Editor's note: Mr. Zou does cost optimization. The specific amount of savings is finally calculated by the financial department. I personally think it is worth learning from. Many companies' cost optimization is calculated by themselves. Lack of credibility makes it difficult for bosses to feel comfortable.
Question: This is an old question. The operation and maintenance team has always stood behind the company's business and is relatively far away from the business. How to better support the business? Or how to explain the value of operation and maintenance to the business? What suggestions do you have?
How to reflect the value specifically? I suggest that if the operation and maintenance team wants to reflect the value, they must first have a service awareness, and then build a service system. The next step is to remain patient and continue to improve. Through this, we can form a positive cycle and make time our friend.
In this section, I will briefly share the guidelines for building a service system for our team. We are customer-centric and build safe, reliable, efficient, low-cost and sustainable services. Output value through service operations, implement service operations through products and tools, and continuously improve. In this guidance outline, we have deeply integrated the three functional roles of operation and maintenance, production and research, and operations in the team. Value is reflected through the output of service operations. Many times, people who work in technology are often not easily aware of the importance of service operations. We often hear people talking about technical operations and product operations, but few people talk about service operations. This has a lot to do with the inertial perception of our technical background. We express more from the perspective of our own professional field, and rarely look at our value from the perspective of the people we serve. When many people mention service, they may simply think of roles such as serving tea and running errands, and they are more repelled by providing service. But in fact, every team is a service team. For example, our service project team serves our end users, and our end users may serve other customers in his field of work. Therefore, providing services is a very important thing. Only by serving customers well and helping them obtain results can you truly reflect your own value.
The above is the detailed content of Tuyou Zou Yi: How to operate and maintain small and medium-sized companies?. For more information, please follow other related articles on the PHP Chinese website!
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
