DataStax Apache Cassandra Developer Associate Certification — Questions and Answers

Question 1: What role does the CommitLog play in Cassandra's write path?

It caches frequently read rows
It provides durability by recording every write before the MemTable (Correct answer)
It indexes partitions for fast lookup
It merges SSTables during compaction

Correct answer: It provides durability by recording every write before the MemTable

The CommitLog is written to first on every mutation to ensure durability in case of a node crash before the MemTable is flushed.

Question 2: What does the term 'eventual consistency' mean in the context of Cassandra?

Reads always return the latest write
All nodes are always in sync
All replicas will converge to the same value given no new updates (Correct answer)
Writes are synchronous across all replicas

Correct answer: All replicas will converge to the same value given no new updates

Eventual consistency guarantees that if no new updates are made, all replicas will eventually return the same value.

Question 3: In Cassandra, what is 'upsert' behavior?

Cassandra merges new and existing rows using a conflict resolution algorithm
Cassandra throws an error if a row already exists during INSERT
Cassandra requires a prior SELECT to check existence before writing
Cassandra always treats INSERT and UPDATE as an upsert — existing rows are overwritten at the cell level (Correct answer)

Correct answer: Cassandra always treats INSERT and UPDATE as an upsert — existing rows are overwritten at the cell level

Cassandra's INSERT and UPDATE both perform cell-level upserts: they write the new value regardless of whether the row exists, with no existence check by default.

Question 4: What is 'key cache' in Cassandra and how does it improve performance?

Caches the entire row for hot partitions
Caches Bloom filter results to avoid recalculation
Caches the most recently executed CQL queries
Caches partition key to SSTable offset mappings, reducing disk seeks for frequently accessed partitions (Correct answer)

Correct answer: Caches partition key to SSTable offset mappings, reducing disk seeks for frequently accessed partitions

The key cache stores the position of a partition key within an SSTable, so Cassandra can skip the index lookup and seek directly to the data on subsequent accesses.

Question 5: What is 'ALLOW FILTERING' in CQL and why should it be used with caution?

It enables full-text search on TEXT columns
It allows filtering on materialized view columns
It enables server-side filtering with secondary indexes only
It forces Cassandra to scan all partitions to satisfy a query without a partition key, which is expensive (Correct answer)

Correct answer: It forces Cassandra to scan all partitions to satisfy a query without a partition key, which is expensive

ALLOW FILTERING permits queries without a partition key filter but causes a full cluster scan, which is extremely inefficient on large datasets.

Question 6: Which compaction strategy is best suited for time-series data that is mostly appended and rarely updated?

UnifiedCompactionStrategy (UCS)
LeveledCompactionStrategy (LCS)
SizeTieredCompactionStrategy (STCS)
TimeWindowCompactionStrategy (TWCS) (Correct answer)

Correct answer: TimeWindowCompactionStrategy (TWCS)

TWCS groups SSTables into time windows and compacts within each window, making it ideal for time-series data with little or no overwrites.

Question 7: In a Cassandra cluster, what does 'bootstrap' refer to?

Restarting all nodes after a cluster-wide upgrade
The process by which a new node joins the ring and streams its assigned data from existing nodes (Correct answer)
Starting the Cassandra daemon for the first time
The initial schema creation when a keyspace is first defined

Correct answer: The process by which a new node joins the ring and streams its assigned data from existing nodes

During bootstrap, a new node contacts the seed nodes, receives its token assignments, and streams the corresponding data from current owners before becoming fully operational.

Question 8: In cassandra.yaml, what does 'concurrent_reads' control?

The number of client connections allowed
The number of threads dedicated to serving read requests (Correct answer)
The CQL read timeout in milliseconds
The maximum number of SSTables read simultaneously

Correct answer: The number of threads dedicated to serving read requests

concurrent_reads sets the size of the thread pool handling read requests; the default is 32 and can be tuned based on CPU cores and workload.

Question 9: What does 'vnodes' (virtual nodes) provide in Cassandra?

Isolated namespaces for keyspaces
Logical shards for multi-tenancy
More even data distribution by assigning multiple token ranges per physical node (Correct answer)
Virtual replicas for read performance

Correct answer: More even data distribution by assigning multiple token ranges per physical node

Vnodes assign multiple non-contiguous token ranges to each physical node, resulting in more balanced data distribution and easier cluster scaling.

Question 10: Which Cassandra compaction strategy minimizes read amplification and is best for mixed read/write workloads?

TimeWindowCompactionStrategy (TWCS)
DateTieredCompactionStrategy (DTCS)
SizeTieredCompactionStrategy (STCS)
LeveledCompactionStrategy (LCS) (Correct answer)

Correct answer: LeveledCompactionStrategy (LCS)

LCS maintains SSTables in levels with size limits, minimizing read amplification to roughly 1 SSTable per read at the cost of higher write I/O.

Question 11: Which CQL keyword allows incrementing or decrementing a numeric column atomically in Cassandra?

UPDATE ... SET col = col + n (counter table) (Correct answer)
INCREMENT
MERGE
ATOMIC

Correct answer: UPDATE ... SET col = col + n (counter table)

Counter tables support `UPDATE tbl SET col = col + 1 WHERE key = ?` for distributed atomic increments/decrements without lightweight transactions.

Question 12: Which CQL function generates a new unique time-based UUID suitable for use as a clustering key?

toTimestamp()
currentTime()
now() (Correct answer)
uuid()

Correct answer: now()

The now() function in CQL generates a new TIMEUUID based on the current time, useful for inserting rows with a time-ordered clustering key.

Question 13: In Cassandra data modeling, what is 'denormalization'?

Partitioning a large table into smaller ones
Storing duplicate data in multiple tables to support different query patterns efficiently (Correct answer)
Normalizing data into separate tables with foreign keys
Removing duplicate columns from a table

Correct answer: Storing duplicate data in multiple tables to support different query patterns efficiently

Because Cassandra cannot perform joins, data is intentionally duplicated across multiple tables, each designed for a specific query access pattern.

Question 14: Which CQL collection type preserves insertion order and allows duplicate values?

list (Correct answer)
map
set
tuple

Correct answer: list

A list is an ordered collection of values that permits duplicates, unlike a set which is unordered and deduplicates, or a map which stores key-value pairs.

Question 15: How does the virtual node (vnode) feature improve cluster operations compared to single-token assignment?

Vnodes reduce the number of SSTables per node
Vnodes compress partition keys to reduce network traffic
Vnodes make data distribution more even and simplify adding or removing nodes (Correct answer)
Vnodes allow a single node to act as both coordinator and replica

Correct answer: Vnodes make data distribution more even and simplify adding or removing nodes

With vnodes, each physical node owns many small token ranges, so bootstrapping and decommissioning redistributes data in smaller, parallel streams.

Question 16: What does 'nodetool move' do in Cassandra?

Moves CommitLog files to a new directory
Transfers ownership of all token ranges to a peer node
Changes the token assignment of the current node to rebalance data distribution (Correct answer)
Migrates all data to a new data center

Correct answer: Changes the token assignment of the current node to rebalance data distribution

nodetool move changes the node's token to a new value, causing Cassandra to stream the appropriate data to/from neighboring nodes to maintain correct distribution.

Question 17: What is the purpose of the 'internode_encryption' setting in cassandra.yaml?

Enforces CQL over SSL for clients
Encrypts CommitLog entries
Enables TLS encryption for communication between Cassandra nodes (Correct answer)
Encrypts data at rest in SSTables

Correct answer: Enables TLS encryption for communication between Cassandra nodes

internode_encryption controls whether TLS is used to secure gossip and streaming traffic between nodes in the cluster.

Question 18: Which cassandra.yaml configuration section controls SSL/TLS encryption for connections between clients and Cassandra nodes?

client_encryption_options (Correct answer)
tls_encryption_options
ssl_options
server_encryption_options

Correct answer: client_encryption_options

client_encryption_options in cassandra.yaml governs TLS settings for the native CQL protocol used by client drivers connecting to Cassandra.

Question 19: What availability risk is introduced by leaving the system_auth keyspace at its default replication factor of 1?

A single node failure can make authentication unavailable cluster-wide (Correct answer)
Performance degrades significantly under concurrent logins
Users will be unable to change their passwords
Superuser credentials become visible to other nodes

Correct answer: A single node failure can make authentication unavailable cluster-wide

With RF=1 for system_auth, if the single node holding a credential replica goes down, authentication requests routed to that token range will fail, potentially locking users out.

Question 20: How do you assign superuser privileges to a role when creating it in Cassandra?

CREATE ROLE admin WITH PRIVILEGE = SUPER
CREATE ROLE admin WITH ROLE = SUPERUSER
CREATE ROLE admin WITH ADMIN = true
CREATE ROLE admin WITH SUPERUSER = true (Correct answer)

Correct answer: CREATE ROLE admin WITH SUPERUSER = true

The SUPERUSER = true option in CREATE ROLE or ALTER ROLE grants full administrative access, allowing the role to manage all other roles and permissions.

Question 21: How do you enable authentication in Cassandra?

Create a superuser via cqlsh only
Enable TLS in cassandra-env.sh
Set authenticator: AllowAllAuthenticator in cassandra.yaml
Set authenticator: PasswordAuthenticator in cassandra.yaml (Correct answer)

Correct answer: Set authenticator: PasswordAuthenticator in cassandra.yaml

Setting authenticator to PasswordAuthenticator in cassandra.yaml enables username/password authentication using the system_auth keyspace.

Question 22: Which of the following data types is utilized to produce our own data kinds?

User-defined data types (Correct answer)
None of the Above
Collection data types
Both Answer

Correct answer: User-defined data types

Explanation: <br> As the name suggests, User-defined data types are used to create data types of our own. In Cassandra, user-defined data types are used to group multiple columns into a single data type, allowing you to define custom structures for your data.

Question 23: What does the CQL statement 'UPDATE ... IF EXISTS' provide?

A batch update across multiple partitions
A lightweight transaction that only applies the update if the row already exists (Correct answer)
An upsert that creates the row if missing
A conditional update using Paxos consensus

Correct answer: A lightweight transaction that only applies the update if the row already exists

IF EXISTS uses Cassandra's lightweight transaction (LWT) mechanism backed by Paxos to perform a compare-and-swap, only updating if the row already exists.

Question 24: What does the 'nodetool drain' command do before a node shutdown?

Clears the key cache and row cache
Decommissions the node and rebalances tokens
Removes the node from the ring
Flushes all MemTables, stops accepting writes, and drains the CommitLog so the node can be safely stopped (Correct answer)

Correct answer: Flushes all MemTables, stops accepting writes, and drains the CommitLog so the node can be safely stopped

nodetool drain ensures all pending writes are flushed to SSTables and the CommitLog is cleared, allowing a clean node shutdown without data loss.

Question 25: What does 'nodetool decommission' do?

Forces a node to restart
Safely removes a node by streaming its data to other nodes before leaving the ring (Correct answer)
Marks a node as down in gossip
Deletes all data on a node immediately

Correct answer: Safely removes a node by streaming its data to other nodes before leaving the ring

nodetool decommission moves the node's token ranges to other nodes and then cleanly removes it from the ring.

Question 26: You need to read the most recent 10 events for a user from a time-ordered table. Which CQL clause should you use?

TOP 10
LIMIT 10 (Correct answer)
ROWNUM <= 10
FETCH FIRST 10 ROWS ONLY

Correct answer: LIMIT 10

The LIMIT clause in CQL restricts the number of rows returned per partition, making it efficient for paginated or top-N queries.

Question 27: Which CQL data type is used to store a universally unique identifier generated natively by Cassandra?

text
uuid
timeuuid (Correct answer)
blob

Correct answer: timeuuid

timeuuid (version 1 UUID) embeds a timestamp, enabling time-ordered sorting and the use of now() and dateOf() functions in CQL.

Question 28: What is a user-defined type (UDT) in Cassandra CQL?

A custom compaction strategy
A custom partition key type
A stored procedure written in Java
A named group of typed fields that can be embedded as a column value in a table (Correct answer)

Correct answer: A named group of typed fields that can be embedded as a column value in a table

A UDT allows you to define a reusable composite type with named fields (e.g., address with street, city, zip) that can be used as a column value in tables.

Question 29: What is the default native transport port used by Cassandra for CQL client connections?

9042 (Correct answer)
7199
7001
7000

Correct answer: 9042

Port 9042 is the default CQL native transport port that drivers and tools like cqlsh use to connect to Cassandra.

Question 30: Which of the subsequent APIs is used by Cassandra's Column Families and Tables?

Column Family-CQL API ;Table-CQL API
Column Family-Thrift API ;Table-Thrift API
Column Family-CQL API ;Table-Thrift API (Correct answer)
Column Family-Thrift API ;Table-CQL API

Correct answer: Column Family-CQL API ;Table-Thrift API

Explanation: <br> Correct Answer: Column Family-CQL API ;Table-Thrift API.

Question 31: What is the difference between a partition key and a clustering key in a Cassandra primary key?

The partition key determines data distribution; clustering keys determine sort order within a partition (Correct answer)
The clustering key determines which node stores data; the partition key sorts rows
There is no difference; they are synonyms
The partition key is optional; the clustering key is mandatory

Correct answer: The partition key determines data distribution; clustering keys determine sort order within a partition

The partition key routes data to a node via consistent hashing, while clustering columns control the physical sort order of rows stored within that partition.

Question 32: Which Cassandra metric indicates that a node is receiving more writes than it can process?

High key cache hit rate
High Bloom filter false positive rate
Low SSTable count
High MutationStage pending tasks in tpstats (Correct answer)

Correct answer: High MutationStage pending tasks in tpstats

A growing pending task count in the MutationStage thread pool indicates write throughput is exceeding the node's processing capacity.

Question 33: What does the `ALLOW FILTERING` clause do in a CQL SELECT statement?

Permits queries without a WHERE clause on primary keys
Forces use of a secondary index
Enables server-side filtering that may require a full partition scan (Correct answer)
Disables client-side filtering for performance

Correct answer: Enables server-side filtering that may require a full partition scan

ALLOW FILTERING permits Cassandra to filter data server-side, potentially scanning many partitions, which can be very slow at scale.

Question 34: What CQL statement is used to create a new keyspace in Cassandra?

CREATE KEYSPACE (Correct answer)
CREATE DATABASE
CREATE NAMESPACE
CREATE SCHEMA

Correct answer: CREATE KEYSPACE

CREATE KEYSPACE is the CQL command to define a new keyspace, specifying the replication strategy and factor.

Question 35: How does increasing the 'memtable_flush_writers' setting improve Cassandra write performance?

It increases the number of threads flushing MemTables to SSTables in parallel, improving throughput on multi-disk setups (Correct answer)
It reduces the MemTable size, causing more frequent smaller flushes
It enables concurrent writes to the same partition
It compresses MemTable data before flushing

Correct answer: It increases the number of threads flushing MemTables to SSTables in parallel, improving throughput on multi-disk setups

More flush writer threads allow Cassandra to flush multiple MemTables to disk concurrently, benefiting systems with multiple data directories or fast SSDs.

Question 36: What does a Cassandra consistency level of QUORUM mean for a cluster with a replication factor of 3?

The nearest replica must acknowledge
At least 2 replicas must acknowledge (Correct answer)
All 3 replicas must acknowledge
Only 1 replica must acknowledge

Correct answer: At least 2 replicas must acknowledge

QUORUM requires a majority, calculated as floor(RF/2) + 1, so for RF=3 that means 2 acknowledgments.

Question 37: What is the 'row cache' in Cassandra and when should it be used?

A cache of CQL prepared statement results
A disk-based cache for recently flushed SSTables
A client-side cache maintained by the driver
An in-JVM cache of entire partitions for tables with very few partitions and high read frequency (Correct answer)

Correct answer: An in-JVM cache of entire partitions for tables with very few partitions and high read frequency

The row cache stores entire partition rows in off-heap memory and is only beneficial for tables with a small number of frequently accessed partitions.

Question 38: A write with consistency level LOCAL_ONE means Cassandra will wait for acknowledgment from:

One replica plus the coordinator node
One replica only in the local datacenter (Correct answer)
The closest replica determined by dynamic snitch
One replica across all datacenters

Correct answer: One replica only in the local datacenter

LOCAL_ONE requires acknowledgment from exactly one replica in the same datacenter as the coordinator, reducing cross-DC latency.

Question 39: Cassandra utilizes _________ by default for cluster communication.

3000
7000 (Correct answer)
5000
9000

Correct answer: 7000

Explanation: <br> Cassandra uses port numbers 9042 for native protocol clients, 7000 for cluster communication (or 7001 if SSL is enabled), and 7199 for JMX by default.

Question 40: What does the 'listen_address' parameter in cassandra.yaml define?

The address of the seed node
The address Cassandra listens on for CQL client connections
The IP address the node uses to communicate with other nodes in the cluster (Correct answer)
The Thrift RPC address

Correct answer: The IP address the node uses to communicate with other nodes in the cluster

listen_address is the IP that other Cassandra nodes use to reach this node for inter-node communication (gossip, streaming, repairs).

Question 41: Which CQL statement correctly grants SELECT permission on table 'orders' in keyspace 'sales' to role 'analyst'?

GRANT READ ON TABLE sales.orders TO analyst
PERMIT SELECT ON TABLE sales.orders TO analyst
ALLOW SELECT ON sales.orders FOR analyst
GRANT SELECT ON TABLE sales.orders TO analyst (Correct answer)

Correct answer: GRANT SELECT ON TABLE sales.orders TO analyst

The correct syntax is GRANT <permission> ON <resource> TO <role>, using standard SQL-style permission names like SELECT, MODIFY, and ALTER.

Question 42: What are the default superuser credentials in a freshly installed Apache Cassandra cluster?

root / cassandra
admin / admin
user / password
cassandra / cassandra (Correct answer)

Correct answer: cassandra / cassandra

Cassandra ships with a default superuser account where both the username and password are 'cassandra', which must be changed immediately in production.

Question 43: What is the purpose of the ALLOW FILTERING clause in a CQL SELECT statement?

Enables server-side post-processing of results not supported by the primary key or index (Correct answer)
Forces a full table scan on all coordinator nodes
Bypasses Cassandra's consistency level check
Allows filtering on static columns

Correct answer: Enables server-side post-processing of results not supported by the primary key or index

ALLOW FILTERING instructs Cassandra to perform an in-memory filter on data that cannot be satisfied by the partition key or a secondary index, which can be expensive.

Question 44: What is 'tombstone' in Cassandra?

A log entry for schema changes
A deletion marker stored in SSTables to represent deleted data (Correct answer)
A backup snapshot file
A marker written at node startup

Correct answer: A deletion marker stored in SSTables to represent deleted data

A tombstone is a deletion marker written to an SSTable that tells Cassandra to ignore older versions of that data during reads.

Question 45: Which JVM garbage collector is recommended for Cassandra to minimize GC pause times?

Serial GC
G1GC or ZGC (Correct answer)
CMS GC
Parallel GC

Correct answer: G1GC or ZGC

G1GC (and increasingly ZGC for newer JDKs) is recommended for Cassandra to keep GC pauses short and avoid stop-the-world events that affect latency.

Question 46: Which authenticator must be configured in cassandra.yaml to require clients to provide a username and password?

PasswordAuthenticator (Correct answer)
AllowAllAuthenticator
SecureAuthenticator
CassandraAuthenticator

Correct answer: PasswordAuthenticator

PasswordAuthenticator stores credentials in the system_auth keyspace and requires clients to supply a username and password during connection.

Question 47: Which directory parameter in cassandra.yaml specifies where SSTables are stored on disk?

commitlog_directory
data_file_directories (Correct answer)
saved_caches_directory
hints_directory

Correct answer: data_file_directories

data_file_directories defines the on-disk path(s) where Cassandra writes SSTable files for all keyspaces and tables.

Question 48: What is the purpose of 'full repair' vs 'incremental repair' in Cassandra?

Full repair compacts all SSTables; incremental repair only compacts new SSTables
Full repair checks all SSTables; incremental repair only checks SSTables written since the last repair (Correct answer)
Full repair runs on the entire cluster; incremental repair runs on a single node
They are equivalent — the terms are interchangeable

Correct answer: Full repair checks all SSTables; incremental repair only checks SSTables written since the last repair

Incremental repair marks repaired SSTables and only re-repairs unrepaired SSTables in subsequent runs, significantly reducing repair time and cluster overhead.

Question 49: What is the recommended approach to add a new datacenter to an existing Cassandra cluster without downtime?

Bootstrap new nodes in the datacenter, then run nodetool rebuild on each to stream data (Correct answer)
Use ALTER KEYSPACE to add the datacenter, then run nodetool repair on existing nodes
Stop all nodes, update cassandra.yaml, then restart the entire cluster
Run nodetool decommission on old datacenter nodes before adding the new ones

Correct answer: Bootstrap new nodes in the datacenter, then run nodetool rebuild on each to stream data

New nodes join the ring, and `nodetool rebuild` streams the required token ranges from an existing datacenter, allowing a live, zero-downtime datacenter addition.

Question 50: What does the gc_grace_seconds property control in a Cassandra table?

The minimum time tombstones are retained before compaction can purge them (Correct answer)
The interval between automatic repair cycles
How long Cassandra waits before evicting data from the memtable
The timeout before a coordinator retries a failed write

Correct answer: The minimum time tombstones are retained before compaction can purge them

gc_grace_seconds (default 864000 = 10 days) defines how long tombstones are kept to ensure deleted data does not resurface on nodes that were down during the deletion.

Question 51: Which architecture model does Apache Cassandra follow?

Master-slave
Hub-and-spoke
Peer-to-peer (masterless) (Correct answer)
Primary-secondary with arbiter

Correct answer: Peer-to-peer (masterless)

Cassandra uses a fully peer-to-peer masterless architecture where every node can accept reads and writes.

Question 52: What is the impact of too many tombstones in a Cassandra partition?

Increased write throughput
Faster compaction cycles
Reduced replication overhead
Degraded read performance because Cassandra must scan and filter tombstones during reads (Correct answer)

Correct answer: Degraded read performance because Cassandra must scan and filter tombstones during reads

Reads must traverse tombstones to determine what data is live; a partition with millions of tombstones can cause read timeouts and TombstoneOverwhelmingException.

Question 53: What is a 'materialized view' in Cassandra?

A view stored in memory only, not persisted to disk
A read-only cached copy of a base table with a different primary key for alternate access patterns (Correct answer)
A writable denormalized copy of a table managed by the user
A cross-keyspace join of two tables

Correct answer: A read-only cached copy of a base table with a different primary key for alternate access patterns

A materialized view is automatically maintained by Cassandra as a separate table built from a base table, allowing efficient queries on non-primary-key columns.

Question 54: What is the purpose of the FROZEN keyword in CQL when used with collections or UDTs?

Compresses the collection on disk
Marks the column as read-only
Enables secondary indexing on nested fields
Serializes the entire collection/UDT as a single blob, preventing partial updates (Correct answer)

Correct answer: Serializes the entire collection/UDT as a single blob, preventing partial updates

FROZEN serializes the entire collection or user-defined type as a single immutable value; updates must replace the entire value, enabling it to be used as a primary key component.

Question 55: What does 'speculative execution' do in Cassandra drivers?

Pre-compiles CQL statements to reduce parse overhead
Sends duplicate requests to multiple replicas and uses the fastest response to reduce tail latency (Correct answer)
Executes a query on a secondary index speculatively
Pre-fetches the next page of results

Correct answer: Sends duplicate requests to multiple replicas and uses the fastest response to reduce tail latency

Speculative execution fires off the same query to an additional replica after a configurable delay, reducing tail latency by using the faster response.

Question 56: Which CQL data type should be used to store a universally unique identifier in Cassandra?

UUID or TIMEUUID (Correct answer)
BLOB
TEXT
BIGINT

Correct answer: UUID or TIMEUUID

UUID stores a random version-4 UUID while TIMEUUID stores a version-1 time-based UUID, both natively supported as primary key types in Cassandra.

Question 57: What does the 'endpoint_snitch' configuration control in Cassandra?

How Cassandra determines the topology (rack/DC) of nodes for replica placement (Correct answer)
The gossip heartbeat interval
The CQL protocol version
Network traffic encryption

Correct answer: How Cassandra determines the topology (rack/DC) of nodes for replica placement

The endpoint_snitch tells Cassandra how nodes are organized into racks and data centers so it can place replicas intelligently to maximize fault tolerance.

Question 58: What does 'hinted handoff' do when a replica node is temporarily down?

The write is permanently rejected
The coordinator waits indefinitely for the down node to recover
Another replica is permanently promoted to own that token range
The coordinator stores the write as a hint and replays it to the node when it recovers (Correct answer)

Correct answer: The coordinator stores the write as a hint and replays it to the node when it recovers

Hinted handoff lets the coordinator temporarily store missed writes for a down replica and deliver them once that node comes back online, improving write availability.

Question 59: What mechanism does Cassandra use to detect and recover from failed nodes automatically?

Paxos consensus
Gossip protocol with hinted handoff (Correct answer)
Raft leader election
ZooKeeper coordination

Correct answer: Gossip protocol with hinted handoff

Cassandra uses gossip for failure detection and hinted handoff to store missed writes temporarily on behalf of a down replica.

Question 60: What is the recommended approach to improve Cassandra read latency for a table that is read in random order with many SSTables?

Trigger nodetool compact to reduce SSTable count (Correct answer)
Switch the consistency level to ONE
Increase the replication factor
Add more clustering columns

Correct answer: Trigger nodetool compact to reduce SSTable count

Reducing the number of SSTables through compaction lowers read amplification since fewer files need to be consulted to reconstruct a partition.

DataStax Apache Cassandra Developer Associate Certification

The DataStax Apache Cassandra Developer Associate certification validates knowledge of CQL, data modeling, Cassandra architecture, and cluster operations. It targets developers and administrators working with Apache Cassandra in production environments.

Questions

90 min

Time Limit

70%

To Pass

Checking for saved progress...

Exam Rules

You can skip questions and return to them later
Flag questions for review before submitting
No feedback shown until you submit the entire exam
Unanswered questions count as wrong — answer everything
10 pretest questions are mixed in and don't affect your score
Timer auto-submits when time runs out
Your progress is auto-saved every 30 seconds