1. Multiple choice questions (10 questions in total, 50 points)
- (Single-choice question) What does a distributed file system refer to?
A. Distribute and store files on multiple computer nodes. Thousands of computer nodes form a computer cluster.
B. Used to transfer data between Hadoop and traditional databases. Delivery
C. A highly available, highly reliable, distributed system for collecting, aggregating and transmitting massive logs
D. A high-throughput distributed publish-subscribe messaging system that can handle all actions in consumer-scale websites Streaming data
My answer: A: Distribute and store files on multiple computer nodes. Thousands of computer nodes constitute a computer cluster.
Correct answer: A: Distribute and store files on multiple computer nodes. Thousands of computer nodes Computer nodes form a computer cluster;
5 points - (Single choice question) Which of the following does not belong to the node in the computer cluster?
A. Master Node (Master Node)
B. Source Node (SourceNode)
C. Name Node (NameNode)
D. Slave Node (Slave Node)
Mine Answer: B: Source Node;
Correct answer: B: Source Node;
5 points - (Single choice question) In HDFS, how big is a block by default?
A. 64MB
B. 32KB
C. 128KB
D. 16KB
My answer: A: 64MB;
Correct answer: A: 64MB;
5 points - (Single-choice question) Which of the following is not a benefit of HDFS adopting the abstract block concept?
A. Simplifies system design
B. Supports large-scale file storage
C. Strong cross-platform compatibility
D. Suitable for data backup
My answer : C: Strong cross-platform compatibility;
Correct answer: C: Strong cross-platform compatibility;
5 points - (Single choice question) In HDFS, what is the main function of NameNode?
A. Maintains the mapping relationship between block id and datanode local files
B. Stores file contents
C. Stores file memory on disk
D. Stores metadata
My answer : A: Maintains the mapping relationship between block id and local file of datanode;
Correct answer: D: Stores metadata;
0 points - (Single choice question) Which of the following descriptions of FsImage is wrong?
A. The FsImage file does not record which data node each block is stored in
. B. The FsImage file contains the serialized form of all directory and file inodes in the file system
. C. FsImage is used to maintain the file system tree and the metadata of all files and folders in the file tree.
D. The FsImage file records the specific data node where each block is stored.
My answer: D: The FsImage file records the specific data node of each block. In which data node is it stored;
Correct answer: D: The FsImage file records which data node each block is stored in;
5 points - (Single choice question) Which of the following descriptions of SecondaryNameNode is wrong?
A. SecondaryNameNode generally runs in parallel on multiple machines
B. It is used to save the backup of HDFS metadata information in the name node , and reduce the time for name node restart
C. SecondaryNameNode obtains FsImage and EditLog files from NameNode through HTTPGET and downloads them to the corresponding local directory
D. SecondaryNameNode is an integral part of the HDFS architecture
. My answer: A: SecondaryNameNode Generally, it runs in parallel on multiple machines;
Correct answer: A: SecondaryNameNode usually runs in parallel on multiple machines;
5 points - (Single choice question) What model does HDFS use?
A. Hierarchical model
B. Master-slave structure model
C. Pipe-filter model
D. Point-to-point model
My answer: B: Master-slave structure model;
Correct answer: B: Master From structural model;
5 points - (Single choice question) In the Hadoop project structure, what does HDFS refer to?
A. Distributed file system
B. Streaming data reading and writing
C. Resource management and scheduler
D. Data warehouse on Hadoop
My answer: A: Distribution Distributed file system;
Correct answer: A: Distributed file system;
5 points - (Single choice question) Which of the following descriptions about HDFS is incorrect?
A. HDFS also adopts corresponding data storage, data reading and data replication strategies to improve the overall read and write response performance of the system
. B. HDFS adopts a master-slave (master-slave) Master/Slave) structural model
C. HDFS uses redundant data storage, which enhances data reliability.
D. HDFS uses the concept of blocks, which makes the system design more complicated.
My answer: D: HDFS uses the concept of blocks, which makes the design of the system more complicated. The design of the system becomes more complex;
correct answer: D: HDFS adopts the concept of blocks, which makes the design of the system more complex;
5 points
2. Multiple choice questions (10 questions in total, 50 points) - (Multiple choice question) Which of the following goals does HDFS want to achieve?
A. Compatibility with cheap hardware devices
B. Streaming data reading and writing
C. Large data sets
D. Complex file models
My answer: ABC: Compatibility with cheap hardware devices; Streaming data reading and writing; Big data set;
Correct answer: ABC: Compatible with cheap hardware devices; Streaming data reading and writing; Big data set;
5 points - (Multiple choice question) The special design of HDFS, while achieving excellent features, also has some application limitations, including which of the following aspects?
A. Poor cross-platform compatibility
B. Inability to efficiently store a large number of small files File
C. Does not support multi-user writing and arbitrary modification of files.
D. Not suitable for low-latency data access.
My answer: BCD: Unable to store a large number of small files efficiently; does not support multi-user writing and arbitrary modification of files; not suitable for low-latency data. Access;
correct answer: BCD: cannot efficiently store a large number of small files; does not support multi-user writing and arbitrary modification of files; is not suitable for low-latency data access;
5 points - (Multiple choice question) Which of the following obvious benefits can HDFS adopt by adopting the abstract block concept?
A. Support large-scale file storage
B. Support small-scale file storage
C. Suitable for data backup
D. Simplify system design
My answer: ACD: supports large-scale file storage; suitable for data backup; simplifies system design;
correct answer: ACD: supports large-scale file storage; suitable for data backup; simplifies system design;
5 points - (Multiple choice question) In HDFS, what core data structures does the NameNode mainly store?
A. FsImage
B. DN8
C. Block
D. EditLog
My answer: AD: FsImage; EditLog;
Correct answer: AD: FsImage; EditLog;
5 points - (Multiple choice question) What are the main functions of the data node (DataNode)?
A. Responsible for the storage and reading of data
B. Storage and retrieval of data according to the scheduling of the client or the name node
C. Regularly sending data to the name node List of blocks stored by itself
D. Used to save the backup of HDFS metadata information in the name node and reduce the time of restarting the name node.
My answer: ABC: responsible for the storage and reading of data; according to the client or name The node schedules to store and retrieve data; regularly sends a list of its stored blocks to the name node;
correct answer: ABC: is responsible for the storage and reading of data; performs data according to the schedule of the client or the name node Storage and retrieval; periodically sends a list of its stored blocks to the namenode;
5 points - (Multiple choice question) What does the HDFS namespace contain?
A. Disk
B. File
C. Block
D. Directory
My answer: BCD: File; Block; Directory;
Correct answer: BCD: File; Block; Directory;
5 points - (Multiple choice question) Which of the following descriptions of the client is correct?
A. The client is the most common way for users to operate HDFS. HDFS provides a client during deployment.
B. The HDFS client is a library that exposes HDFS File system interface
C. Strictly speaking, the client is not part of HDFS
D. The client can support common operations such as opening, reading, and writing.
My answer: ABCD: The client is the most common way for users to operate HDFS. , HDFS provides a client when deployed; the HDFS client is a library that exposes the HDFS file system interface; strictly speaking, the client is not part of HDFS; the client can support opening, reading, writing, etc. Common operations;
correct answer: ABCD: The client is the most common way for users to operate HDFS. HDFS provides a client when deployed; the HDFS client is a library that exposes the HDFS file system interface; strictly speaking, the client Not considered part of HDFS; the client can support common operations such as opening, reading, and writing;
5 points - (Multiple choice question) HDFS only sets up a unique name node. Although this greatly simplifies the system design, it also brings some obvious limitations?
A. Namespace restrictions
B. Performance bottlenecks
C. Isolation issues
D. Availability of the cluster
My answer: ABCD: Namespace limitations; Performance bottlenecks; Isolation issues; Cluster availability;
Correct answer: ABCD: Namespace limitations; Performance bottlenecks; Isolation issues; Cluster availability;
5 points - (Multiple choice question) Which of the following advantages does HDFS data block multi-copy storage have?
A. Speeds up data transmission
B. Easily detects data errors
C. Ensures data reliability
D. Suitable for running on multiple platforms
My answer: ABC: Speeds up data transmission Speed; easy to check data errors; ensure data reliability;
correct answer: ABC: speed up data transmission; easy to check data errors; ensure data reliability;
5 points - (Multiple choice question) HDFS has high fault tolerance. What corresponding mechanisms have been designed to detect data errors and perform automatic recovery?
A. The data source is too large
. B. The data node is wrong
. C. The data is wrong.
D. The name node is wrong.
My answer: BCD: Data node error; Data error; Name node error;
Correct answer: BCD