The original intention of this book:
Any book is a carrier for carrying and transmitting knowledge, and readers can explore the answers they want from it. For me, books are the fastest way to take me through the wonderful world of computers. The reason why I want to create a book related to large-scale website architecture is because I have experienced too many technical problems in the actual development process in recent years. Whenever my technical team and I try to solve these problems, they will first Try to find a solution from the existing technical books on the market; but things are counterproductive. At present, most of the books on the architecture theory of Gaoge are on the market, and there are very few books that really explain the architecture solutions of large-scale websites. Regarding the gaps in this field, I want to try to create and write down my experience and experience over the years so that more people can benefit. After all, the architecture needs to be implemented, otherwise it will be empty talk.
Content explanation
There are 5 chapters in this book. Almost all the contents of "Mu Yi Cao" are independent, so you can read them selectively.
Chapter 1 Big System and Small Do-Large-scale Service Architecture
Chapter 1 takes the large-scale service-oriented architecture as the opening chapter of the book. It mainly introduces the evolution process of the distributed system architecture and how to implement service governance in the large-scale service invocation scenario.
1.1 The evolution of the architecture of the distributed system
The Internet has quietly changed the world, changed people's perception of things, and shortened the distance between people. Whether you want to admit it or not, the Internet has completely affected and integrated into our lives. The author's mother has never been a person who likes to catch up with the trend, but she has never left her smart device for a long time. The first thing to get up every morning is to pick up her smart phone, scan the circle of friends, check current affairs and politics, and do it back." "Eating melon people" gossip and entertainment news, even food, clothing, housing and transportation are almost all done through the Internet, as shown in Figure 1-1. Since the Internet can make our lives better, please embrace it tightly with open arms.
Chapter 2 Big Promotion to Prepare Nuclear Bombs-Full Link Pressure Test
Chapter 2 focuses on how to implement full-link stress testing online, as well as instructive capacity planning and performance optimization on the eve of the big promotion, to make the system as solid as a rock.
2.1 Why do we need to implement full-link stress testing online
Before introducing the theme of this chapter, please calmly think about what things we need to consider on the eve of the big promotion? Or what things must be done, as much as possible, and do not fight unprepared battles. The author summarized the two most basic and difficult preparation tasks on the eve of the big promotion:
- Evaluate the number of machine expansions,
- Verify whether the overall system capacity can effectively support the estimated peak traffic.
Chapter 3 Xiao Tiangu-Flow Control Program
Chapter 3 focuses on how to effectively control the flow. If a reasonable and effective method is used to control the peak flow and make it access the system in an orderly manner, the system can operate stably under any circumstances.
3.1 Why current limit is needed
Before discussing why the system needs to limit current, let's talk about the flow control scenarios that can be seen everywhere in our lives. The author's place of residence and work are both in Shenzhen. Since it is a first-tier city, take the subway as an example when traveling. During the rush hours of working days, the subway station can be described as overcrowded. During this period, the load pressure of the subway station is simply worse than that during the Spring Festival. It originally only takes about 5 minutes to get from the station hall to the platform. However, under the traffic control of subway security personnel, it took 20 to 30 minutes to enter the platform smoothly, which was more than five times the usual. I believe students who squeezed through the bus and subway should feel the same.
Chapter 4 Big Promotion to Buy Core Technology Betting Questions-Read/Write Optimization Scheme
Chapter 4 focuses on how to solve core technical problems such as high concurrency and high concurrency in the context of big sales promotion.
4.1 Introduction to caching technology
Cache is no longer a new technology. In the actual development process, almost all development students have dealt with it. In short, caching refers to storing frequently accessed hotspot data in the closest place to computing to facilitate the system to respond quickly, such as static resource data (including pictures, audio, video, script files and HTML web pages, etc.) ), we can cache it on the CDN (Content Delivery Network), because the user’s request does not go to the company’s data center, but to the nearest ISP (Internet Service Provider) to the user Therefore, the overall response speed of the system can be greatly improved, as shown in Figure 4-1.
Chapter 5 Spreading Around-Sub-database and Sub-table Scheme
Chapter 5 introduces the evolution of the relational database architecture in detail, and also focuses on how to ensure the final consistency of the data in the actual order business scenario.
5.1 The evolution of relational database architecture
In the Internet scenario, there are two main performance bottlenecks common to databases, as shown below: a large number of concurrent read/write operations, resulting in unbearable load pressure on a single database; excessive storage of data in a single table, resulting in low retrieval efficiency .
Friends who need to get this "Everyone is an Architect 2.0" version can directly forward + follow the private message (learning) to get it for free
Service governance needs
With the increase in business complexity, servicing can effectively help companies solve the problems of repeated construction of shared services, horizontal scaling of business systems, and large-scale business development team collaboration. Then the author will focus on large-scale servicing for everyone How to implement service governance under the scenario.
The call chain of service governance
As shown in Figure 1-21, in a large-scale service invocation scenario, the dependencies between services can be described as intricate, and even architects cannot sort out the dependencies and calling sequence between services in a short time.
Friends who need to get this "Everyone is an Architect 2.0" version can directly like + follow "Add me VX assistant" to get it for free
