I have been in the industry for more than ten years, and I have grown from an interviewer yearning for a big factory to a big factory interviewer. Many
of my friends who are engaged in Java development often consult how to enter a big factory. Indeed, for most people, entering a large factory is not just a talk of capital, but a technological pursuit.Large companies like Ali, Tencent, and Byte have significantly better technical atmosphere and technical specifications than smaller companies, and their business scenarios are richer and more challenging. A few years of training in a large factory can better improve personal ability and have more choices for future career planning.
Having said that, it is not easy to get an offer from a major factory. Resume delivery is like a big wave, and the interview process is also fierce. In the special period, the demand of large factories decreases, but the requirements are higher. Without thorough preparation, it is difficult to pass through layers of screening. The problem with many Java people is not that they are not technically inadequate, but that they have not formed a systematic and comprehensive knowledge system and are always speechless when asked by interviewers.
Combining my many years of experience as interviewers in major e-commerce companies, I sorted out the knowledge points frequently asked over the years and organized them into 100 core interview questions for major e-commerce companies, covering major test points, high-frequency points and salary increases. , I believe it will be helpful for everyone to prepare for the interview, and it can also solve some of the problems encountered by developers in daily projects.
These points are divided into 5 categories, namely: Dubbo, ElasticSearch, JVM, multithreading/high concurrency, and message middleware. I have disassembled and thought about these issues into a pdf. Due to space limitations, I will not list them here.
I interviewed many Internet companies and took comprehensive notes. I think it is quite good. Now I will share it with you and give you a reference. Due to limited space, only a small part of the interview questions are shown below. Friends who need the full version can click a link to jump to receive: link: just click! ! ! Password: CSDN
Here is a list of some technical assessment points in the past three years, specifically:
1 Dubbo
1.1 How to solve the service call timeout problem?
1.2 What serialization methods does Dubbo support?
1.3 The relationship between Dubbo and SpringCloud?
1.4 Dubbo's architecture design? What are the levels divided?
1.5 Dubbo's default cluster fault tolerance solution?
1.6 What communication framework does Dubbo use?
1.7 The main application scenarios of Dubbo?
1.8 Dubbo service registration and discovery process? Flow Description.
1.9 What are Dubbo's cluster fault tolerance solutions?
1.10 The four major components of
Dubbo 1.11 How does
Dubbo solve the security mechanism 1.12 The difference between Dubbo and SpringCloud?
1.13 Which protocols does Dubbo support, the application scenarios of each protocol, the advantages and disadvantages?
1.14 What are the core functions of Dubbo?
1.15 Dubbo's registry cluster is down. Can publishers and subscribers still communicate?
1.16 What are the strategies for load balancing in Dubbo cluster
1.17 Why is service governance needed?
1.18 How to set Dubbo timeout time?
2 ElasticSearch
2.1 In your company's ES cluster, how many shards are generally allocated to a node?
2.2 How does Elasticsearch implement Master election?
2.3 How do you do write tuning?
2.4 How to avoid split brain?
2.5 How does 19-Elasticsearch realize the aggregation of large amounts of data (on the order of hundreds of millions)?
2.6 Can the number of ES main shards be changed later? why?
2.7 How to monitor the cluster status?
2.8 What is a copy in ElasticSearch?
2.9 20. ES update data execution process?
2.10 What constitutes a shard?
2.11 What is the analyzer in ElasticSearch?
2.12 What is split brain?
2.13 How does the client select a specific node to execute the request when connecting to the cluster?
2.14 What is the inverted index in Elasticsearch?
2.15 What is an index? Index (noun) An index (index)
2.16 Describe in detail the process of Elasticsearch updating and deleting documents
3 JVM
3.1 JVM parameters mainly have several classifications
3.2 3. Will there be memory leaks in Java, briefly describe.
3.3 How does the Java virtual machine determine that two Java classes are the same?
3.4 What reference types are there
in Java 3.5 When can an object be garbage collected in Java?
3.6 19. Have you encountered StackOverflow exceptions? Under what circumstances do you guess it will be triggered?
3.7 What are the parts of heap space? And how to set each part?
3.8 What is a stack frame? What is stored in the stack frame?
3.9 How to set parameters to generate GC log?
3.10 What is GC? Why is there a GC?
3.12 Which jdk commands have been used, and explain what each function is
3.13 What are the parts of the data area during JVM runtime?
3.14 Do you understand the mechanism of the parent delegation model of the class loader and destroy the parent delegation model?
3.15 How many types of escape analysis are there?
3.16 What is the meaning of these parameters -Xms?
3.17 Do you know which kinds of garbage collectors, their advantages and disadvantages, focus on cms and G1, including principles, processes, advantages and disadvantages.
3.18 What is the ratio of Eden and Survivor in the memory structure of JVM?
4 Multithreading/high concurrency
4.1 What is the significance of load balancing?
4.2 Please tell me how to synchronize threads and thread scheduling?
4.3 Which statement is correct about the difference between epoll and select? (Multiple choice)
A. Both epoll and select are I/O multiplexing technologies, both of which can monitor the status of multiple I/O events at the same time.
B. epoll is more efficient than select, mainly based on the I/O event notification mechanism supported by its operating system, while select is based on the polling mechanism.
C. epoll supports two modes: horizontal trigger and edge trigger.
D. select can support I/O in parallel, which is relatively small and cannot be modified.
4.4 Does starting a thread call the run() method or the start() method?
4.5 How to ensure that N threads can access N resources without causing deadlock?
4.6 How to write a multi-threaded program (another question: how to create a multi-threaded program)?
4.7 The difference between thread and process?
4.8 What is a thread pool and what are the commonly used thread pools?
4.9 What is a deadlock?
4.10 How to ensure the consistency of cache and database data?
5 message middleware
5.1 How many modes are there for consumers to obtain information?
5.2 17. What are the characteristics of RocketMQ?
5.3 Kafka has set 7 days and 10G to clear data at the same time. By the fifth day, the message reached 10G. How will Kafka handle this?
5.4 Why a Kafka cluster is needed
5.5 Kafka data storage design
5.6 How does Kafka judge whether a node is alive?
5.7 There are several reliability mechanisms for sending Kafka messages.
5.8 Please elaborate on the push mode and pull mode.
5.9 There are three key differences between Kafka and traditional messaging systems
5.10 What are the roles of RocketMQ?
5.12 23. How do
Kafka consumers consume data 5.13 Advantages of
Kafka 5.14 What is the design of Kafka?
5.15 What do you know about Consumer?
5.16 In which directory will Kafka create new partitions
5.17 Talk about the consumption process of Kafka consumers
5.18 Introduce Kafka
5.19 What will cause Kafka to run slower?
These points are divided into 5 categories, namely: Dubbo, ElasticSearch, JVM, multithreading/high concurrency, and message middleware. I have disassembled and thought about these issues and sorted them into a pdf. Due to space limitations, I will not list them all here.