MCQs on Advanced Data Modeling | Cassandra
Chapter 7 of Cassandra focuses on advanced data modeling techniques, essential for optimizing data storage and access in large-scale applications. This chapter includes concepts such as secondary indexes, materialized views, collections and UDTs, time-series data modeling, and designing for high throughput and scalability. These strategies are key to maximizing Cassandra’s performance.
Secondary Indexes
- What is the purpose of a secondary index in Cassandra?
a) To store the data in a compressed format
b) To improve query performance on non-primary key columns
c) To replicate data across clusters
d) To manage user authentication - Secondary indexes in Cassandra are useful for:
a) Efficiently querying by primary key
b) Reducing the number of nodes in the cluster
c) Querying on non-primary key columns
d) Encrypting sensitive data - Which of the following is true about secondary indexes in Cassandra?
a) They can be used for all data types
b) They work by creating a global index across all nodes
c) They should only be used for low-cardinality data
d) They are always faster than primary key queries - A drawback of secondary indexes in Cassandra is:
a) Increased write latency
b) Improved read performance
c) Reduced replication overhead
d) Simplified schema design - When should you avoid using secondary indexes in Cassandra?
a) For high-cardinality columns
b) For small datasets
c) For frequently updated columns
d) For low-cardinality columns
Materialized Views
- What is a materialized view in Cassandra?
a) A copy of data stored in a compact format
b) A precomputed query result stored as a table
c) A table optimized for fast updates
d) A feature used to manage database backups - Materialized views in Cassandra are used for:
a) Storing historical data
b) Querying data with a different primary key
c) Optimizing network latency
d) Reducing the number of nodes - Which of the following is true about materialized views in Cassandra?
a) They are manually maintained
b) They can create new tables with different primary keys
c) They cannot be queried directly
d) They automatically manage replication - When a materialized view is updated in Cassandra, what also gets updated?
a) The original base table
b) Only the indexes
c) The replication factor
d) The SSTables - One of the limitations of materialized views in Cassandra is:
a) High write latency
b) Limited compatibility with all data types
c) They do not support indexing
d) Data redundancy and consistency issues
Using Collections and UDTs
- What is a collection in Cassandra?
a) A group of rows with the same column value
b) A set of predefined values in a single column
c) A type of partition key
d) A distributed query execution plan - Which data types are supported as collections in Cassandra?
a) List, Set, Map
b) String, Integer, Float
c) Date, Timestamp, Blob
d) Tuple, UDT, Binary - What is a User-Defined Type (UDT) in Cassandra?
a) A schema-less data type
b) A custom data structure for grouping related fields
c) A primary key for the table
d) A secondary index - When using collections in Cassandra, you can store:
a) Large blobs only
b) Sets and lists of primitive types
c) Only scalar values
d) Single-value columns only - Which operation can be performed on collections in Cassandra?
a) Performing arithmetic operations
b) Nested indexing
c) Modifying individual elements within collections
d) Encrypting the entire collection
Time-Series Data Modeling
- What is a common use case for time-series data modeling in Cassandra?
a) User authentication management
b) Storing logs and metrics
c) Replicating data across multiple nodes
d) Managing access control - In time-series data modeling, what is a typical approach for creating a partition key?
a) Use a combination of time and sensor ID
b) Use the geographic location
c) Use the user ID
d) Use random values for better distribution - What is the ideal data structure in Cassandra for time-series data?
a) Wide rows with timestamps as clustering keys
b) Small partitions with frequent updates
c) Static rows with precomputed aggregates
d) Large tables with single-column values - A common challenge in time-series data modeling in Cassandra is:
a) Too many write operations
b) Managing schema updates
c) Efficiently storing high volumes of data
d) Handling queries on non-primary key columns - To avoid “hotspots” in time-series data modeling in Cassandra, you should:
a) Use a single partition key
b) Use wide rows with time as the clustering key
c) Store all data in one large table
d) Avoid indexing time-based data
Designing for High Throughput and Scalability
- To achieve high throughput in Cassandra, you should:
a) Avoid using secondary indexes
b) Use consistent read and write consistency levels
c) Use heavy clustering of nodes
d) Focus on small tables and simple queries - Which technique improves scalability in Cassandra?
a) Using a monolithic architecture
b) Distributing data across multiple nodes with partitioning
c) Relying on a master-slave architecture
d) Centralizing data storage in one node - What is a key principle for designing scalable Cassandra applications?
a) Use centralized clusters for fast queries
b) Distribute read and write loads across multiple nodes
c) Ensure all nodes have identical data
d) Limit partitioning to small datasets - What is the effect of high write throughput on Cassandra?
a) It reduces system uptime
b) It improves query performance
c) It may increase write latency if not optimized
d) It causes data to be deleted automatically - In Cassandra, how can you optimize for high throughput?
a) Use large amounts of RAM for each node
b) Ensure data is evenly distributed and avoid hotspots
c) Minimize the number of read operations
d) Use complex queries to reduce CPU load - How does data partitioning in Cassandra improve scalability?
a) It allows for large tables to be compressed
b) It spreads data across nodes to prevent overload on a single node
c) It simplifies query processing by using a single node
d) It reduces the amount of data replication required - What is a good practice for managing large datasets in Cassandra for scalability?
a) Keep data in large partitions
b) Avoid writing data in bulk
c) Optimize partition size for fast writes
d) Use frequent schema changes - What happens if you do not design for scalability in Cassandra?
a) Data redundancy is increased
b) The system will fail under heavy loads
c) Data is stored in a non-redundant format
d) Clustering becomes inefficient - To maximize throughput in a Cassandra cluster, it is important to:
a) Avoid replication across data centers
b) Use simple queries with low complexity
c) Ensure minimal data duplication
d) Perform complex join operations - Which of the following is an example of a scalability strategy in Cassandra?
a) Vertical scaling by adding more CPU cores
b) Horizontal scaling by adding more nodes to the cluster
c) Reducing the number of nodes in the cluster
d) Adding a single, centralized node
Answers
| Qno | Answer |
|---|---|
| 1 | b) To improve query performance on non-primary key columns |
| 2 | c) Querying on non-primary key columns |
| 3 | c) They should only be used for low-cardinality data |
| 4 | a) Increased write latency |
| 5 | a) For high-cardinality columns |
| 6 | b) A precomputed query result stored as a table |
| 7 | b) Querying data with a different primary key |
| 8 | b) They can create new tables with different primary keys |
| 9 | a) The original base table |
| 10 | d) Data redundancy and consistency issues |
| 11 | b) A set of predefined values in a single column |
| 12 | a) List, Set, Map |
| 13 | b) A custom data structure for grouping related fields |
| 14 | b) Sets and lists of primitive types |
| 15 | c) Modifying individual elements within collections |
| 16 | b) Storing logs and metrics |
| 17 | a) Use a combination of time and sensor ID |
| 18 | a) Wide rows with timestamps as clustering keys |
| 19 | c) Efficiently storing high volumes of data |
| 20 | b) Use wide rows with time as the clustering key |
| 21 | b) Use consistent read and write consistency levels |
| 22 | b) Distributing data across multiple nodes with partitioning |
| 23 | b) Distribute read and write loads across multiple nodes |
| 24 | c) It may increase write latency if not optimized |
| 25 | b) Ensure data is evenly distributed and avoid hotspots |
| 26 | b) It spreads data across nodes to prevent overload on a single node |
| 27 | c) Optimize partition size for fast writes |
| 28 | b) The system will fail under heavy loads |
| 29 | b) Use simple queries with low complexity |
| 30 | b) Horizontal scaling by adding more nodes to the cluster |