Courseiva

AWS Certified Database Specialty DBS-C01 (DBS-C01) — Questions 15761650

1663 questions total · 23pages · All types, answers revealed

Page 21

Page 22 of 23

Page 23
1576
MCQmedium

A company runs an e-commerce application on Amazon RDS for MySQL. The application experiences read-heavy traffic during flash sales. The development team wants to offload read traffic without changing the application code. Which solution should be used?

A.Implement an Amazon ElastiCache cluster and update the application to cache queries.
B.Configure Multi-AZ deployment for the RDS instance.
C.Create an RDS Read Replica and point read traffic to the replica endpoint.
D.Use DynamoDB Accelerator (DAX) in front of the RDS instance.
AnswerC

Read Replicas offload read traffic without application changes.

Why this answer

Creating an RDS Read Replica allows read-heavy traffic to be offloaded from the primary RDS instance without any application code changes. The application simply needs to be configured to use the read replica's endpoint for SELECT queries, while writes continue to the primary instance. This directly addresses the requirement to offload read traffic without modifying the application code.

Exam trap

The trap here is that candidates often confuse Multi-AZ with read scaling, assuming the standby instance can serve reads, but in RDS Multi-AZ the standby is not accessible for read traffic—it only provides failover redundancy.

How to eliminate wrong answers

Option A is wrong because implementing ElastiCache requires updating the application code to cache queries, which violates the requirement of no code changes. Option B is wrong because Multi-AZ deployment provides high availability and automatic failover, but does not offload read traffic; the standby instance cannot serve reads. Option D is wrong because DynamoDB Accelerator (DAX) is an in-memory cache for Amazon DynamoDB, not for RDS for MySQL, and cannot be placed in front of an RDS instance.

1577
MCQhard

A company is migrating an on-premises Oracle database to Amazon RDS for Oracle. The database is 2 TB in size and has a high transaction rate. The migration must have minimal downtime. Which migration strategy should a database specialist recommend?

A.Use AWS DMS with ongoing replication to capture changes.
B.Copy the data files to Amazon S3 and then load into RDS.
C.Use Oracle Data Pump to export and import the database.
D.Perform an RMAN backup and restore to RDS.
AnswerA

DMS with CDC allows near-zero downtime migration.

Why this answer

AWS DMS with ongoing replication allows for minimal downtime by continuously replicating changes from the source Oracle database to the target Amazon RDS for Oracle, allowing a cutover with only a brief pause. Option B is incorrect because copying data files to S3 and loading into RDS is not a supported migration path for RDS for Oracle and would require significant downtime. Option C is incorrect because Oracle Data Pump export and import requires the source database to be offline or in read-only mode during export, causing downtime.

Option D is incorrect because RMAN backup and restore is not natively supported for RDS for Oracle and would involve complex manual steps and downtime.

1578
MCQmedium

A database administrator notices that an Amazon Aurora MySQL DB cluster's primary instance is experiencing high CPU utilization due to a long-running analytical query. The application can tolerate eventual consistency. What is the MOST cost-effective solution to reduce CPU load on the primary instance?

A.Configure an Auto Scaling policy for Aurora Replicas and modify the application to use the reader endpoint.
B.Create a read replica in a different region and route read traffic there.
C.Increase the instance size of the primary instance.
D.Enable Aurora Parallel Query on the cluster.
AnswerA

Aurora Replicas share the same storage and can offload SELECT queries.

Why this answer

Offloading read queries to Aurora Replicas is cost-effective and reduces primary load without additional infrastructure.

1579
MCQeasy

A team is migrating an on-premises database to Amazon RDS using AWS DMS. After the full load completed, the task stopped. The team needs to perform ongoing replication to minimize downtime. What change should the team make to the DMS task?

A.Modify the replication task to change the migration type to 'full-load-and-cdc'.
B.Restart the replication task with the same settings.
C.Create a new replication instance and start a new full-load task.
D.Use AWS Schema Conversion Tool to convert the schema again.
AnswerA

Changing to 'full-load-and-cdc' enables both full load and ongoing change data capture replication.

Why this answer

After a full load completes and the task stops, the team needs to enable ongoing replication to capture and apply changes made to the source database during the migration window. Modifying the DMS task to use the 'full-load-and-cdc' migration type allows the task to perform the initial full load and then seamlessly transition to Change Data Capture (CDC) to replicate incremental changes, minimizing downtime. This is the correct approach because DMS supports switching a completed full-load task to include CDC without re-running the full load.

Exam trap

The trap here is that candidates may think restarting the task or creating a new instance will automatically enable CDC, but DMS requires explicitly modifying the migration type to include CDC after the full load completes.

How to eliminate wrong answers

Option B is wrong because restarting the replication task with the same settings will simply re-run the full load from scratch, not enable ongoing replication. Option C is wrong because creating a new replication instance and starting a new full-load task would duplicate effort and waste resources, and it still does not enable CDC for ongoing replication. Option D is wrong because the AWS Schema Conversion Tool (SCT) is used for schema conversion, not for enabling CDC or managing replication tasks; it is irrelevant to the need for ongoing replication.

1580
MCQeasy

A startup is building a mobile app that requires a scalable NoSQL database. The data model includes user profiles with variable attributes that change over time. The database must support high read throughput and low latency. Which AWS database is best suited?

A.Amazon Neptune
B.Amazon RDS for MySQL
C.Amazon DynamoDB
D.Amazon Redshift
AnswerC

NoSQL, flexible schema, high performance at scale.

Why this answer

Amazon DynamoDB is a fully managed NoSQL key-value and document database that delivers single-digit millisecond latency at any scale. It supports flexible schema with variable attributes, making it ideal for user profiles that change over time, and its provisioned or on-demand capacity modes enable high read throughput with consistent low latency.

Exam trap

The trap here is that candidates may confuse Amazon Neptune's graph capabilities with NoSQL flexibility, or assume a relational database like MySQL can handle variable attributes via JSON columns, overlooking DynamoDB's native schema-less design and guaranteed single-digit millisecond performance at scale.

How to eliminate wrong answers

Option A is wrong because Amazon Neptune is a graph database designed for highly connected data (e.g., social networks, fraud detection), not for general-purpose NoSQL workloads with variable attributes. Option B is wrong because Amazon RDS for MySQL is a relational database with a fixed schema, requiring predefined columns and table alterations for attribute changes, which contradicts the variable-attribute requirement. Option D is wrong because Amazon Redshift is a petabyte-scale data warehouse optimized for analytical queries (OLAP), not for high-throughput, low-latency transactional reads (OLTP) on user profiles.

1581
Multi-Selecthard

A company is running an Amazon RDS for MySQL Multi-AZ DB instance. The company wants to implement a backup strategy that meets a recovery point objective (RPO) of 5 minutes and a recovery time objective (RTO) of 1 hour. Which THREE steps should the company take to meet these objectives?

Select 3 answers
A.Configure AWS Backup to copy automated backups to another region.
B.Enable automated backups with a retention period of 35 days.
C.Change the DB instance to a Single-AZ configuration to reduce costs.
D.Create a cross-region read replica to use for failover in another region.
E.Take daily manual snapshots and copy them to another region.
AnswersA, B, D

AWS Backup can copy backups cross-region for disaster recovery.

Why this answer

Automated backups with a 35-day retention period (B) enable point-in-time recovery within the last 35 days, supporting an RPO of 5 minutes. Configuring AWS Backup to copy automated backups to another region (A) provides cross-region disaster recovery, helping meet the RTO of 1 hour by restoring in a different region. Creating a cross-region read replica (D) allows for quick failover by promoting the replica, which can significantly reduce RTO.

Option C (Single-AZ) is less resilient and does not help meet RPO/RTO. Option E (daily manual snapshots) cannot achieve a 5-minute RPO due to manual timing.

1582
Multi-Selecthard

A company is migrating a 5 TB Microsoft SQL Server database to Amazon Aurora MySQL. The migration must be completed with minimal downtime. Which THREE steps should the company take? (Select THREE.)

Select 3 answers
A.Update the application connection string to point to the Aurora reader endpoint.
B.Use AWS DMS to perform a full load and ongoing replication from SQL Server to Aurora.
C.Create a read replica of the Aurora database in the same region.
D.Use AWS SCT to convert the SQL Server schema to MySQL-compatible schema.
E.Update the application connection string to point to the Aurora writer endpoint after cutover.
AnswersB, D, E

DMS handles data migration with minimal downtime.

Why this answer

AWS DMS (Database Migration Service) supports ongoing replication (change data capture) from Microsoft SQL Server to Amazon Aurora MySQL, enabling a full load followed by continuous sync of changes. This minimizes downtime by allowing the source database to remain operational until the final cutover, at which point the application switches to the Aurora writer endpoint.

Exam trap

The trap here is that candidates often confuse the Aurora reader endpoint (for read replicas) with the writer endpoint (for primary writes), and incorrectly assume a read replica is needed during migration, when in fact the migration requires schema conversion and ongoing replication via DMS.

1583
MCQmedium

A developer needs to connect to the RDS instance from an EC2 instance in the same VPC. The EC2 instance's security group allows outbound traffic to 0.0.0.0/0. The RDS security group inbound rules currently allow traffic from 0.0.0.0/0 on port 3306. After a security review, the company decides to restrict inbound traffic to the VPC only. Which inbound rule should be added to the RDS security group?

A.An inbound rule allowing traffic from the RDS endpoint address on port 3306.
B.An inbound rule allowing traffic from the EC2 instance's security group ID on port 3306.
C.An inbound rule allowing traffic from 10.0.0.0/8 on port 3306.
D.An inbound rule allowing traffic from the VPC CIDR (e.g., 10.0.0.0/16) on port 3306.
AnswerB

This restricts access to only the EC2 instances in that security group.

Why this answer

Referencing the EC2 instance's security group ID in the RDS inbound rule is the most secure and specific method to restrict access to only that EC2 instance within the VPC. Option A is incorrect because the RDS endpoint is a DNS name for the database instance, not a source IP or security group. Option C is too broad, allowing any traffic from the 10.0.0.0/8 range, which may include subnets outside the VPC.

Option D is less specific than using the security group ID and could permit access from any resource within the VPC CIDR, whereas the security group approach provides granular control.

1584
MCQhard

A financial services company is migrating an on-premises Oracle database to AWS. The database supports an OLTP application with complex joins, stored procedures, and requires high availability within a single Region. The company wants to minimize licensing costs and use a fully managed service. Which AWS database service should they choose?

A.Amazon DynamoDB
B.Amazon RDS for Oracle
C.AWS Database Migration Service (DMS)
D.Amazon Aurora PostgreSQL-Compatible Edition
AnswerD

Aurora PostgreSQL is fully managed, supports complex joins and stored procedures, and provides high availability with Multi-AZ.

Why this answer

Amazon Aurora PostgreSQL-Compatible Edition is the correct choice because it is a fully managed, high-availability database service that supports complex joins, stored procedures, and OLTP workloads while minimizing licensing costs. Aurora provides built-in replication across three Availability Zones, automatic failover, and up to 15 read replicas, meeting the high availability requirement without the licensing overhead of commercial databases like Oracle.

Exam trap

The trap here is that candidates may choose Amazon RDS for Oracle (Option B) because it supports Oracle features directly, overlooking the explicit requirement to minimize licensing costs and the fact that Aurora PostgreSQL can handle complex joins and stored procedures without Oracle licensing fees.

How to eliminate wrong answers

Option A is wrong because Amazon DynamoDB is a NoSQL key-value and document database that does not support complex joins, stored procedures, or the relational schema required by the existing Oracle OLTP application. Option B is wrong because Amazon RDS for Oracle would require purchasing Oracle licenses (Bring Your Own License or included license), which contradicts the goal of minimizing licensing costs, and it is not the most cost-effective fully managed option for high availability. Option C is wrong because AWS Database Migration Service (DMS) is a migration tool, not a database service; it helps move data to AWS but does not provide the operational database or high availability itself.

1585
Multi-Selectmedium

Which TWO actions can help reduce Amazon RDS for MySQL replication lag between a primary instance and a read replica? (Choose two.)

Select 2 answers
A.Increase the allocated storage for the read replica.
B.Reduce the number of write-heavy DML statements on the primary.
C.Enable Multi-AZ on the primary instance.
D.Increase the instance size of the read replica.
E.Disable binary logging on the primary instance.
AnswersB, D

Fewer changes to replicate means less lag.

Why this answer

Reducing write-heavy DML on the primary (Option B) decreases the volume of changes to be replicated, directly reducing lag. Increasing the instance size of the read replica (Option D) provides more CPU and memory resources for applying changes from the binary log. Option A is incorrect because increasing allocated storage does not improve the replica's ability to apply changes; it only increases storage capacity.

Option C is incorrect because enabling Multi-AZ on the primary provides high availability but does not reduce replication lag. Option E is incorrect because binary logging is required for replication; disabling it would break replication.

1586
Multi-Selecthard

Which TWO steps are required when migrating an on-premises Microsoft SQL Server database to Amazon RDS for SQL Server using AWS DMS with ongoing replication?

Select 2 answers
A.Create a VPC peering connection between the on-premises network and the DMS VPC.
B.Install the DMS agent on the source SQL Server instance.
C.Enable MS-CDC (change data capture) on the source database.
D.Configure MS-Replication on the source database.
E.Ensure the source database is using the full recovery model.
AnswersC, E

DMS uses CDC to capture ongoing changes from SQL Server.

Why this answer

AWS DMS requires MS-CDC (change data capture) to be enabled on the source SQL Server database when using ongoing replication (change data capture). CDC captures insert, update, and delete activity from the transaction log, allowing DMS to apply those changes to the target RDS for SQL Server instance without requiring full table reloads.

Exam trap

The trap here is that candidates often confuse MS-CDC with MS-Replication, assuming DMS requires replication agents or topology, when in fact DMS uses native CDC or log-based capture without setting up replication.

1587
MCQmedium

A company runs a critical application on Amazon RDS for PostgreSQL with a Multi-AZ deployment. The application experiences intermittent connection timeouts and slow query performance. The CloudWatch metrics show that the 'ReadLatency' and 'WriteLatency' metrics are elevated during peak hours. The 'CPUUtilization' is consistently below 30%, and 'DatabaseConnections' is within limits. The 'BurstBalance' for the gp2 storage is frequently dropping to 0%. The DB instance is a db.r5.large with 300 GB of gp2 storage. The company wants to resolve the latency issues without significant cost increase. Which solution should the company implement?

A.Add a read replica to offload read traffic.
B.Enable Performance Insights to identify the root cause.
C.Switch the storage type to io1 with 3000 provisioned IOPS.
D.Increase the allocated storage to 600 GB to increase baseline IOPS.
AnswerD

Larger gp2 volumes have higher baseline IOPS, reducing burst credit depletion.

Why this answer

Increasing the allocated storage to 600 GB increases the baseline IOPS of gp2 from 900 (at 300 GB) to 1800 (at 600 GB), reducing reliance on burst credits and resolving the latency issue. This is cost-effective compared to switching to io1 (Option C), which would incur higher costs for 3000 provisioned IOPS. Option A (adding a read replica) does not help with write latency, and Option B (enabling Performance Insights) only aids diagnosis, not resolution.

1588
MCQhard

A company is using Amazon Aurora MySQL-Compatible Edition. The security team wants to ensure that database credentials are not stored in application configuration files. They decide to use AWS Secrets Manager to manage credentials. The application is hosted on Amazon EC2 instances that have an IAM role attached. What is the most secure way to grant the application access to the secret?

A.Grant the IAM role attached to the EC2 instance permissions to read the secret, and use the Secrets Manager API to retrieve it at runtime.
B.Store the secret in the application code and rotate it periodically.
C.Attach a resource-based policy to the EC2 instance allowing access to the secret.
D.Store the secret in an encrypted S3 bucket and have the application download it at startup.
AnswerA

This is the most secure approach because the secret is never stored on the instance; it is retrieved on demand via API.

Why this answer

The most secure approach is to grant the EC2 IAM role permission to read the secret using an IAM policy attached to the role, and then have the application call the Secrets Manager API to retrieve the secret at runtime. This avoids storing secrets in code or configuration files. Option B is wrong because storing the secret in application code is insecure.

Option C is wrong because resource-based policies are not attached to EC2 instances; IAM roles are the correct mechanism. Option D is wrong because storing secrets in an S3 bucket introduces additional complexity and potential exposure; using Secrets Manager with IAM roles is more secure.

1589
MCQmedium

A company stores sensitive data in an Amazon S3 bucket that is accessed by an Amazon Redshift cluster. The security team requires that the data in transit between Redshift and S3 be encrypted. Which configuration ensures this?

A.Enable server-side encryption (SSE-S3) on the S3 bucket.
B.Ensure that the Redshift cluster has SSL enabled, which encrypts data in transit by default for COPY/UNLOAD operations to S3.
C.Configure a VPC endpoint for S3.
D.Use client-side encryption on the data before uploading to S3.
AnswerB

Redshift uses SSL for data transfer to S3 by default.

Why this answer

Redshift automatically uses SSL encryption for data in transit when moving data to and from S3 using COPY/UNLOAD commands. Option A is incorrect because server-side encryption (SSE-S3) encrypts data at rest, not in transit. Option C is incorrect because client-side encryption encrypts data before sending, but Redshift handles this automatically with SSL.

Option D is incorrect because VPC endpoints do not encrypt data in transit; they provide private connectivity.

1590
Multi-Selecthard

A company is using Amazon DynamoDB with on-demand capacity mode. The application experiences occasional throttling during traffic spikes. The database administrator wants to implement a solution to reduce throttling. Which THREE actions should the administrator take? (Choose THREE.)

Select 3 answers
A.Use DynamoDB Accelerator (DAX) to cache frequently accessed items.
B.Enable auto scaling for the table.
C.Switch the table to provisioned capacity mode with auto scaling.
D.Implement exponential backoff and retry logic in the application.
E.Design the application to handle throttling by retrying with backoff.
AnswersA, D, E

DAX reduces read traffic to the table, decreasing the chance of read throttling.

Why this answer

Options A, D, and E are correct. Using DynamoDB Accelerator (DAX) caches frequently accessed items, offloading read capacity from the table and reducing read throttling. Implementing exponential backoff and retry logic in the application (option D) and designing the application to handle throttling by retrying with backoff (option E) both ensure that requests are retried gracefully during spikes, reducing the impact of throttling.

Option B is incorrect because auto scaling is not applicable to on-demand capacity mode, which scales automatically. Option C is incorrect because switching to provisioned capacity does not inherently reduce throttling; on-demand already handles sudden spikes, but throttling may still occur if the workload exceeds table limits momentarily.

1591
MCQmedium

A company is migrating a PostgreSQL database to Amazon RDS for PostgreSQL. The source database uses custom functions written in PL/Python. What should the company consider?

A.Rewrite the functions in PL/pgSQL or another supported language.
B.Modify the DB parameter group to enable PL/Python.
C.Use RDS Custom for PostgreSQL to install PL/Python.
D.Migrate to Aurora PostgreSQL which supports PL/Python.
AnswerA

RDS PostgreSQL supports only specific procedural languages; PL/Python is not one of them.

Why this answer

Amazon RDS for PostgreSQL does not support PL/Python because it requires the installation of an untrusted procedural language, which poses security risks in a managed database environment. The correct approach is to rewrite the custom functions in a supported language such as PL/pgSQL, PL/Perl, or PL/Tcl, which are trusted languages available in RDS for PostgreSQL.

Exam trap

The trap here is that candidates assume RDS Custom or Aurora PostgreSQL can bypass the PL/Python limitation, but AWS does not support untrusted languages in any managed PostgreSQL offering, and RDS Custom still relies on the standard PostgreSQL engine which lacks PL/Python support.

How to eliminate wrong answers

Option B is wrong because modifying the DB parameter group cannot enable PL/Python; RDS for PostgreSQL does not include the PL/Python extension and does not allow installation of untrusted languages. Option C is wrong because RDS Custom for PostgreSQL still does not support PL/Python; RDS Custom provides OS-level access but the PostgreSQL engine itself must support the language, and PL/Python is not bundled or supported by AWS. Option D is wrong because Aurora PostgreSQL also does not support PL/Python; it shares the same limitations as RDS for PostgreSQL regarding untrusted procedural languages.

1592
Multi-Selecteasy

Which TWO AWS services can be used together to migrate an on-premises MySQL database to Amazon Aurora MySQL with minimal downtime? (Select TWO.)

Select 2 answers
A.AWS Schema Conversion Tool (SCT)
B.Amazon Route 53
C.AWS CloudFormation
D.AWS Database Migration Service (DMS)
E.AWS Global Accelerator
AnswersA, D

Converts schema for compatibility.

Why this answer

AWS Database Migration Service (DMS) can perform continuous replication from an on-premises MySQL source to an Amazon Aurora MySQL target using change data capture (CDC), enabling near-zero downtime. The AWS Schema Conversion Tool (SCT) is used to assess and convert the source schema, including any incompatible objects, to ensure the Aurora target is ready for migration. Together, SCT handles schema transformation while DMS handles the ongoing data replication.

Exam trap

The trap here is that candidates often assume a single service like DMS alone can handle schema conversion, but DMS does not convert incompatible schema objects — that is the specific role of SCT, and both are required for a complete migration with minimal downtime.

1593
MCQhard

Refer to the exhibit. A company has an S3 bucket with server-side encryption using AWS KMS. The bucket policy allows access to a user, but the user receives 'Access Denied' when trying to download an object. The IAM policy for the user is shown. What is the MOST likely cause?

A.The user does not have MFA enabled.
B.The KMS key ID in the policy is incorrect.
C.The policy does not allow kms:Decrypt for the S3 bucket.
D.The policy has a condition that limits kms:Decrypt to when the request is made through the S3 service, but the user might be using a different service to access the object (e.g., directly via KMS API).
AnswerD

The condition 'kms:ViaService' ensures decryption only through S3.

Why this answer

The IAM policy includes a condition that restricts kms:Decrypt to requests made through the S3 service (via the S3 service principal). If the user attempts to decrypt the object directly using the KMS API (e.g., using AWS CLI with `kms decrypt` command), the condition is not met, resulting in an 'Access Denied' error. Option A is incorrect because MFA is not mentioned in the policy.

Option B is incorrect because the KMS key ID is assumed correct as per the scenario. Option C is incorrect because the policy does allow kms:Decrypt, but only conditionally.

1594
MCQhard

A company uses Amazon DynamoDB with fine-grained access control based on user identity. The IAM policy for a user includes a condition that restricts access to items where the 'department' attribute equals 'engineering'. However, the user is receiving an 'AccessDeniedException' when trying to query items. What is the most likely cause?

A.The user is not using the correct IAM role.
B.The 'department' attribute is not the partition key of the table.
C.The DynamoDB table is not configured to require SSL.
D.The IAM policy is missing a 'Condition' block.
AnswerB

Fine-grained access control using 'LeadingKeys' requires the condition on the partition key.

Why this answer

Fine-grained access control in DynamoDB uses IAM policy conditions with the 'dynamodb:LeadingKeys' or 'dynamodb:Attributes' condition keys. The 'department' attribute must be the partition key of the table for 'LeadingKeys' to work. If 'department' is not the partition key, the condition cannot be evaluated for the query.

Option A is possible but less likely if the user is using the correct role. Option C is wrong because DynamoDB does not require SSL for all operations. Option D is wrong because Condition check is exactly how it works.

1595
Multi-Selecthard

A company runs a web application on Amazon EC2 instances behind an Application Load Balancer. The application uses Amazon RDS for MySQL. Recently, the database CPU utilization spikes to 100% during peak hours. The team observes that the spike is caused by a large number of slow queries. They need to identify and resolve the issue with minimal disruption. Which combination of steps should they take? (Choose two.)

Select 2 answers
A.Enable RDS Performance Insights to identify the slow queries
B.Upgrade to a larger instance type
C.Increase the DB instance storage to improve I/O
D.Create a read replica and direct reporting queries to the replica
E.Implement connection pooling using Amazon RDS Proxy
AnswersA, D

Performance Insights provides a dashboard to analyze database performance and identify problematic queries.

Why this answer

The correct combination is A and D. Option A: Enabling RDS Performance Insights quickly identifies slow queries and their resource consumption, pinpointing the cause of CPU spikes. Option D: Creating a read replica and directing reporting queries to it offloads read traffic from the primary instance, reducing CPU load during peak hours.

Option B is not an identification step and may not address the root cause, while also incurring unnecessary cost. Option C addresses I/O bottlenecks, not CPU spikes from slow queries. Option E helps with connection management but does not identify or directly resolve the slow query issue.

1596
MCQeasy

A company is using Amazon DynamoDB with on-demand capacity. The application experiences increased latency during peak hours. The DynamoDB table receives about 5,000 writes per second and 10,000 reads per second. The read latency is high, but write latency is acceptable. What should the company do to reduce read latency?

A.Enable DynamoDB Streams to offload read traffic.
B.Create a read replica of the DynamoDB table.
C.Enable DynamoDB Accelerator (DAX) for the table.
D.Switch to provisioned capacity and enable auto scaling.
AnswerC

DAX provides an in-memory cache that reduces read latency for DynamoDB tables.

Why this answer

DynamoDB Accelerator (DAX) is a fully managed, highly available, in-memory cache that significantly reduces read latency for DynamoDB tables, especially for repeated reads. DAX allows you to offload read traffic from the table and serve read requests from the cache, reducing the load on the table and improving response times. Option A is wrong because DynamoDB Streams is used to capture changes to items in the table, not to cache data or offload read traffic.

Option B is wrong because DynamoDB does not support read replicas; DAX is the caching solution for reads. Option D is wrong because switching to provisioned capacity and enabling auto scaling would help with scaling write capacity but does not directly address high read latency; DAX is specifically designed for in-memory caching to improve read performance.

1597
MCQhard

A healthcare company stores patient records in Amazon DynamoDB. Each record includes patient_id (partition key), visit_date (sort key), and a large JSON attribute for medical history. The application frequently queries recent visits for a patient and scans historical data for analytics. The scans on the medical history attribute cause high RCU consumption. The company wants to reduce costs and improve query performance. Which design should be implemented?

A.Compress the medical history attribute using gzip before storing in DynamoDB.
B.Move the medical history attribute to a separate table with patient_id as partition key and visit_date as sort key. Use DynamoDB Streams to keep both tables in sync.
C.Enable DynamoDB Accelerator (DAX) for the table to cache frequent queries.
D.Use Amazon S3 to store the medical history as a separate object and reference it from DynamoDB.
AnswerB

Separating the large attribute reduces RCU consumption for queries that do not need it.

Why this answer

It separates the large, infrequently accessed medical history attribute from the frequently queried core record, reducing the item size for common queries and thus lowering RCU consumption. By using DynamoDB Streams to synchronize the two tables, you maintain data consistency without adding complexity to the application, and queries against the main table become faster and cheaper since they no longer read the large JSON payload.

Exam trap

The trap here is that candidates often choose compression (Option A) thinking it reduces storage and read costs, but DynamoDB does not natively support compression and charges based on the actual stored item size, so compression must be handled at the application layer and does not reduce RCU consumption.

How to eliminate wrong answers

Option A is wrong because compressing the medical history attribute with gzip before storing it in DynamoDB does not reduce RCU consumption; DynamoDB charges for the actual stored size of the item, and compression is not transparent to read operations—the application would still need to read the compressed data and decompress it, and the item size remains the same from DynamoDB's perspective. Option C is wrong because enabling DAX caches query results but does not reduce the RCU cost of the initial scan or query; DAX is a cache layer that speeds up repeated reads but does not change the fact that scanning the large medical history attribute consumes high RCUs per request. Option D is wrong because while storing medical history in S3 and referencing it from DynamoDB is a valid pattern, it introduces latency for retrieving the history and requires additional application logic to fetch the S3 object; more importantly, it does not address the high RCU consumption from scans on the DynamoDB table itself, as the scans would still need to read the reference attribute (which is small) but the question specifically targets the scans on the medical history attribute causing high RCU consumption.

1598
MCQhard

A company runs an OLTP workload on an RDS for MySQL instance. The database has a table with 50 million rows. The application frequently runs queries that join this table with a small lookup table (1000 rows) using a foreign key. The queries are slow. Which design change would most improve performance?

A.Partition the large table by the foreign key column.
B.Scale up the RDS instance to a larger size.
C.Add an index on the foreign key column in the large table.
D.Create a read replica and direct all read queries to it.
AnswerC

An index on the join column allows the database to quickly find matching rows, dramatically improving join performance.

Why this answer

The query joins a large table (50M rows) with a small lookup table (1000 rows) on a foreign key column. Without an index on the foreign key column in the large table, MySQL must perform a full table scan for each join, leading to slow performance. Adding an index on that column allows MySQL to use an index lookup (e.g., B-tree) to quickly locate matching rows, dramatically reducing query time.

Exam trap

The DBS-C01 exam often tests the misconception that partitioning or read replicas can fix join performance issues, but the real bottleneck is typically a missing index on the join column in the large table.

How to eliminate wrong answers

Option A is wrong because partitioning the large table by the foreign key column does not inherently speed up joins; it only splits data into physical segments, and queries still need to scan relevant partitions unless partition pruning is applied, which is not guaranteed for join conditions. Option B is wrong because scaling up the RDS instance increases CPU and memory but does not address the root cause of missing index; the query will still perform full table scans, wasting resources. Option D is wrong because creating a read replica and directing read queries to it does not improve join performance; the replica still lacks the necessary index, so queries remain slow on the replica.

1599
MCQeasy

A startup is deploying a new web application on AWS. The application uses Amazon RDS for PostgreSQL as its database. The developers need to create the initial schema and load test data. They have a SQL script file that creates tables and inserts 10,000 rows. The DB instance is a db.t3.micro with 20 GB gp2 storage. The developers connect to the RDS instance using pgAdmin from their local machine and run the script. The script fails with a 'connection reset' error after inserting 5,000 rows. What is the MOST likely cause?

A.The DB instance ran out of memory or CPU credits due to the intensive script
B.The max_connections parameter was set too low
C.The storage volume reached its maximum capacity
D.The security group blocked the connection due to a timeout
AnswerD

Correct. A security group can have timeout settings that cause the connection to be reset if the script runs longer than the timeout, which is a common issue with long-running queries through clients like pgAdmin.

Why this answer

The most likely cause is that the connection was terminated by an idle timeout on an intermediate network component (such as a NAT Gateway, which has a default idle timeout of 350 seconds, or the client-side firewall/proxy). Security groups are stateful and do not have configurable timeout settings, so they are not directly responsible. The script's duration likely exceeded the idle timeout, causing a 'connection reset' error.

Option A is unlikely because a db.t3.micro can handle inserting 10,000 rows without exhausting memory or CPU credits; the error is network-related, not resource-related. Option B is incorrect because max_connections is sufficient for a single connection. Option C is incorrect because 20 GB storage is far more than needed for 10,000 rows.

1600
Multi-Selectmedium

A company is migrating a 500 GB PostgreSQL database from on-premises to Amazon Aurora PostgreSQL. They need to minimize downtime and ensure that all data changes made during migration are synchronized. Which three components are required for this migration? (Choose THREE.)

Select 3 answers
A.AWS Database Migration Service (DMS) replication instance
B.AWS Schema Conversion Tool (SCT)
C.Amazon Aurora PostgreSQL cluster as the target
D.Source PostgreSQL database configured with logical replication
E.AWS VPN connection between on-premises and AWS
AnswersA, C, D

DMS orchestrates the migration and replication.

Why this answer

AWS DMS replication instance is required because it orchestrates the continuous replication of data from the source PostgreSQL database to the target Aurora PostgreSQL cluster. It manages the connection, reads changes from the source using logical replication slots, and applies them to the target, enabling near-zero downtime migration.

Exam trap

The DBS-C01 exam often tests the misconception that AWS Schema Conversion Tool is always needed for any database migration, but it is only required when migrating between different database engines, not for homogeneous migrations like PostgreSQL to Aurora PostgreSQL.

1601
MCQhard

A company runs a critical application on Amazon RDS for PostgreSQL. The application team reports that the database occasionally becomes unresponsive for a few seconds. CloudWatch metrics show 'CPUSurplusCreditsCharged' and 'CPUSurplusCredits' are not 0. The instance is a db.t3.medium. What is the likely cause and how should it be fixed?

A.The instance is out of CPU credits and is being throttled; switch to a larger or non-burstable instance
B.Enable Enhanced Monitoring to diagnose the issue
C.Increase the allocated storage to improve I/O
D.The instance is experiencing a failover; enable Multi-AZ
AnswerA

Correct because T3 instances are burstable; if CPU credits are exhausted, the instance is throttled. Switching to a non-burstable or enabling unlimited mode resolves the issue.

Why this answer

T3 instances are burstable and use CPU credits. When credits are exhausted, they can use surplus credits, which incur charges. If the workload is consistently high, the instance may run out of credits and become throttled, causing unresponsiveness.

The fix is to switch to T3 unlimited mode (but that incurs charges) or use a non-burstable instance (e.g., M5). Option B is wrong because Enhanced Monitoring provides detailed OS-level metrics but does not prevent CPU credit exhaustion. Option C is wrong because increasing allocated storage does not affect CPU credits; it improves I/O performance.

Option D is wrong because enabling Multi-AZ provides high availability and failover but does not resolve CPU credit issues.

1602
MCQhard

A financial services company runs an Amazon Aurora MySQL database. The application performs complex joins and aggregations on large tables, causing high CPU utilization on the writer instance. The team wants to reduce load without changing the application code. Which solution would best address this issue?

A.Create one or more Aurora Replicas and route read traffic to them.
B.Switch to asynchronous replication to reduce load on the primary.
C.Enable Aurora Auto Scaling to increase storage capacity.
D.Migrate to Aurora Serverless v2 for automatic scaling.
AnswerA

Read replicas offload SELECT queries from the writer.

Why this answer

Creating one or more Aurora Replicas and routing read traffic to them offloads the complex joins and aggregations from the writer instance, reducing CPU utilization without requiring application code changes. Aurora Replicas share the same underlying storage volume as the writer, so they serve read queries with minimal replication lag while the writer focuses on write operations.

Exam trap

The trap here is that candidates may confuse scaling compute capacity (Aurora Serverless v2) with offloading read traffic, but only read replicas directly reduce CPU load on the writer by moving read-heavy operations to separate instances.

How to eliminate wrong answers

Option B is wrong because switching to asynchronous replication does not reduce CPU load on the primary; it only changes how data is replicated to replicas, and Aurora already uses asynchronous replication between the writer and replicas. Option C is wrong because Aurora Auto Scaling for storage capacity increases storage automatically but does not offload compute or reduce CPU utilization on the writer instance. Option D is wrong because migrating to Aurora Serverless v2 provides automatic scaling of compute capacity but does not inherently separate read and write workloads; the writer instance would still handle all complex queries, so CPU load would remain high.

1603
MCQeasy

A company is migrating an on-premises Oracle database to Amazon RDS for Oracle. The database contains sensitive data that must be encrypted at rest. Which action should the company take to enable encryption on the RDS instance?

A.Modify the existing RDS DB instance and enable encryption.
B.Create a new RDS DB instance with encryption enabled and migrate the data.
C.Enable Oracle Transparent Data Encryption (TDE) on the existing RDS instance.
D.Take a snapshot of the existing database, encrypt the snapshot, and restore from it.
AnswerB

Encryption can only be enabled at creation time.

Why this answer

Amazon RDS for Oracle supports encryption at rest using AWS KMS. Encryption can only be enabled when the DB instance is created; you cannot encrypt an existing unencrypted RDS instance. Therefore, to enable encryption for an existing on-premises Oracle database being migrated, you must create a new RDS DB instance with encryption enabled at launch and then migrate the data into that new instance.

This makes option B the correct choice. Option A is incorrect because you cannot modify an existing unencrypted RDS instance to add encryption. Option C is incorrect because while Oracle Transparent Data Encryption (TDE) can be used to encrypt data within the database, it is not the native Amazon RDS encryption method and requires additional configuration; moreover, the scenario assumes a new RDS instance is being created, not an existing one.

Option D is incorrect because the source database is on-premises, not an RDS instance, so there is no RDS snapshot to take; even if there were an existing unencrypted RDS instance, encrypting a snapshot requires copying it with encryption enabled, not directly encrypting the snapshot. Thus D is not a valid action for this migration.

1604
MCQmedium

A company is running an Amazon RDS for MySQL Multi-AZ DB instance. The primary instance in us-east-1a experiences an unexpected failure. After the automatic failover, the application team reports that write latency has increased significantly. The new primary instance is in us-east-1b. The DB instance class and storage configuration are identical. What is the MOST likely cause of the increased write latency?

A.The DB instance class in us-east-1b is a different size than the original.
B.The application is connecting to the DB instance in a different Availability Zone, increasing network latency.
C.The Multi-AZ configuration uses asynchronous replication, causing higher latency.
D.The new primary is in the same Availability Zone as the application, but the standby is in a different AZ, causing synchronous replication overhead.
AnswerB

The new primary is in us-east-1b, and if the application is in us-east-1a, cross-AZ latency increases write latency.

Why this answer

After failover, the new primary DB instance resides in us-east-1b, while the application likely continues to connect to the original endpoint or is still running in us-east-1a. This cross-AZ network hop introduces additional latency for write operations, as the application must send data over the network between Availability Zones. The DB instance class and storage are identical, so performance differences are not due to hardware changes.

Exam trap

The trap here is that candidates may assume Multi-AZ failover is transparent and does not affect performance, but cross-AZ network latency for writes can increase significantly if the application remains in the original AZ.

How to eliminate wrong answers

Option A is wrong because the question explicitly states that the DB instance class and storage configuration are identical, so a size difference is not a factor. Option C is wrong because Multi-AZ replication is synchronous, not asynchronous; asynchronous replication would be used for cross-Region read replicas, not for Multi-AZ failover. Option D is wrong because the new primary is in us-east-1b, not the same AZ as the application (which is in us-east-1a), and synchronous replication overhead applies to the standby, not to the application's write path.

1605
Multi-Selecthard

A company's Amazon Aurora MySQL DB cluster is experiencing a failover event. Which THREE metrics in CloudWatch should be examined to understand the cause of the failover?

Select 3 answers
A.ACUUtilization
B.ReadLatency
C.BinLogDiskUsage
D.DatabaseConnections
E.FailoverCount
AnswersA, D, E

High ACU utilization can trigger failover.

Why this answer

(ACUUtilization) is correct because high ACU utilization can indicate resource exhaustion (CPU/memory pressure) that may trigger a failover in Aurora. Option D (DatabaseConnections) is correct because a sudden spike or drop in connections could signal issues like connection exhaustion or application failures that lead to failover. Option E (FailoverCount) is correct because this metric directly tracks the number of failover events; examining the timestamp of the last increment helps correlate with the incident.

Option B (ReadLatency) is wrong because while high read latency can be a symptom, it is not a direct cause of failover; Aurora uses different mechanisms for read scaling. Option C (BinLogDiskUsage) is wrong because it relates to binary log storage for replication, not a direct trigger for failover.

1606
MCQmedium

A company is using Amazon RDS for MySQL with automated backups enabled. The security team requires that all backups be encrypted at rest. Which configuration ensures that new automated backups are encrypted?

A.Create a new encrypted DB instance and migrate the data. Automated backups will inherit encryption.
B.Modify the DB instance to enable encryption at rest after creation.
C.Use AWS KMS to encrypt the S3 bucket where automated backups are stored.
D.Enable encryption at rest on the DB instance and also enable encryption on automated backups using the DBSnapshot option.
AnswerA

Creating an encrypted instance ensures all backups (automated and manual) are encrypted.

Why this answer

Encryption at rest must be enabled at creation time on an RDS instance. When you enable encryption at rest on a new DB instance, all automated backups, snapshots, and replicas inherit that encryption automatically. Option B is incorrect because you cannot enable encryption at rest on an existing RDS instance after creation; you must create a new encrypted instance and migrate.

Option C is incorrect because automated backups are stored in an S3 bucket managed by RDS, but you cannot directly encrypt the bucket or backups separately; encryption is tied to the DB instance. Option D is incorrect because it incorrectly suggests both enabling encryption on the instance and a separate DBSnapshot option; encryption on the instance already covers backups, and there is no separate DBSnapshot option for enabling encryption.

1607
MCQmedium

A company uses Amazon Redshift for data warehousing. They run a query that joins a large fact table (10 billion rows) with a small dimension table (1 million rows). The query is slow. The distribution style of the fact table is AUTO, and the dimension table has DISTSTYLE ALL. The join key is user_id. What is the MOST likely reason for the poor performance?

A.The dimension table does not have a sort key on user_id
B.The fact table's distribution key is not user_id, causing redistribution
C.The dimension table uses DISTSTYLE ALL, which is inefficient for joins
D.The fact table should have column compression disabled for the join key
AnswerB

AUTO may distribute by another key, leading to large data movement during join.

Why this answer

When the fact table uses DISTSTYLE AUTO, Redshift may choose a distribution key that is not user_id. When the fact table is distributed on a different key, joining on user_id requires Redshift to redistribute the fact table rows across nodes to match the dimension table's distribution, causing significant network traffic and slower performance. The dimension table with DISTSTYLE ALL is already replicated to all nodes, so the bottleneck is the fact table's distribution mismatch.

Exam trap

The trap here is that candidates often assume DISTSTYLE ALL is always inefficient for joins, but in this scenario it is actually beneficial, while the real culprit is the fact table's distribution key not matching the join key due to AUTO assignment.

How to eliminate wrong answers

Option A is wrong because sort keys optimize data ordering for range-restricted scans and merging, not for join redistribution; the slow join is due to data movement, not sorting. Option C is wrong because DISTSTYLE ALL is actually efficient for small dimension tables in joins, as it replicates the table to all nodes, avoiding redistribution of the dimension table. Option D is wrong because disabling column compression on the join key would increase I/O and storage costs without addressing the redistribution overhead; compression does not affect join performance in this context.

1608
MCQhard

A company uses Amazon RDS for SQL Server with Multi-AZ deployment. During a failover test, the application experienced a longer downtime than expected. Which monitoring metric should be reviewed to understand the failover duration?

A.FailoverTime
B.WriteLatency
C.DatabaseConnections
D.ReplicaLag
AnswerC

DatabaseConnections reflects the number of active connections. During a failover, connections are lost and then re-established. The period of low or zero connections approximates the failover duration, making it the best metric among the options.

Why this answer

The DatabaseConnections metric in Amazon CloudWatch tracks the number of active database connections. During a Multi-AZ failover, existing connections are dropped and new connections are blocked until the standby becomes the primary. By monitoring the time period when DatabaseConnections drops to zero or a low value, you can estimate the failover duration.

Options A (FailoverTime) is not a valid CloudWatch metric for RDS; the actual metric is 'Failover' which is a count, not a duration. WriteLatency may increase during failover but does not directly measure downtime. ReplicaLag applies to read replicas, not Multi-AZ failover.

Exam trap

Beware of non-existent metrics. 'FailoverTime' sounds plausible but is not a CloudWatch metric. The actual metric is 'Failover' (count).

1609
Multi-Selecthard

Which THREE factors should be considered when choosing between a homogeneous migration and a heterogeneous migration to Amazon RDS? (Select THREE.)

Select 3 answers
A.Network latency between source and target.
B.Performance requirements of the target database.
C.Amount of application code changes needed.
D.Schema compatibility between source and target.
E.Licensing costs of the target database engine.
AnswersB, D, E

May influence engine choice.

Why this answer

A homogeneous migration (e.g., MySQL to RDS MySQL) typically requires minimal schema changes, preserves performance characteristics, and maintains existing licensing costs. A heterogeneous migration (e.g., Oracle to RDS PostgreSQL) requires schema conversion (validating schema compatibility), may require re-architecting to meet performance requirements due to different query optimizers, and can change licensing costs (e.g., eliminating Oracle licenses). Therefore, factors B, D, and E directly influence the choice between homogeneous and heterogeneous migrations.

Exam trap

The trap here is that candidates confuse migration method factors (e.g., network latency, downtime tolerance) with migration type factors (e.g., schema compatibility, licensing, performance), leading them to select A as a relevant consideration when it only applies to the migration approach, not the choice between homogeneous and heterogeneous.

1610
Multi-Selecteasy

A database administrator is monitoring an Amazon RDS for PostgreSQL DB instance. The administrator notices that the DB instance is using more memory than expected. Which TWO metrics in Amazon CloudWatch can help diagnose memory usage?

Select 2 answers
A.NetworkReceiveThroughput
B.FreeableMemory
C.ReadIOPS
D.DatabaseConnections
E.SwapUsage
AnswersB, E

This metric shows the amount of available RAM.

Why this answer

The correct answers are FreeableMemory (B) and SwapUsage (E). FreeableMemory reflects the amount of available memory on the instance; a low value indicates high memory usage. SwapUsage tracks the amount of swap space being used; when memory is exhausted, the OS begins swapping, which directly indicates memory pressure.

The other metrics are unrelated to memory: NetworkReceiveThroughput (A) measures network traffic, ReadIOPS (C) measures I/O operations, and DatabaseConnections (D) shows the number of connections, not memory consumption.

1611
MCQhard

A security engineer is designing a VPC with an RDS instance. The database must not be accessible from the internet, but EC2 instances in a private subnet must connect. Which security group configuration is MOST secure?

A.Set the RDS security group inbound rule to allow traffic from the VPC CIDR on port 3306.
B.Set the RDS security group inbound rule to allow traffic from 0.0.0.0/0 on port 3306.
C.Set the RDS security group inbound rule to allow traffic from the private subnet CIDR on port 3306.
D.Set the RDS security group inbound rule to allow traffic from the EC2 security group on port 3306.
AnswerD

This is the most secure, least-privilege approach.

Why this answer

The most secure because it uses a security group reference instead of a CIDR block. By referencing the EC2 security group as the source in the RDS security group inbound rule, only instances associated with that specific security group can communicate with the database on port 3306. This approach adheres to the principle of least privilege and automatically scales as instances are added or removed from the EC2 security group, without needing to update CIDR ranges.

Exam trap

The trap here is that candidates often choose a CIDR-based rule (like the private subnet CIDR) thinking it is sufficiently restrictive, but they overlook the superior security and operational benefits of using a security group reference, which is a key concept tested in the DBS-C01 exam for database security.

How to eliminate wrong answers

Option A is wrong because allowing traffic from the entire VPC CIDR (e.g., 10.0.0.0/16) is overly permissive; any resource in the VPC, including those in public subnets or unintended services, could reach the RDS instance, increasing the attack surface. Option B is wrong because allowing traffic from 0.0.0.0/0 on port 3306 makes the database publicly accessible from the internet, directly violating the requirement that the database must not be accessible from the internet. Option C is wrong because while it restricts traffic to the private subnet CIDR, it still allows any instance in that subnet (including compromised or unauthorized ones) to connect, and it does not automatically adjust if the subnet CIDR changes or if instances are added to a different private subnet.

1612
Multi-Selecthard

A company runs a MySQL-compatible database on Amazon RDS for a mission-critical application. The database experiences high write latency due to frequent index updates. The team wants to redesign the database to reduce write amplification and improve insert performance. Which TWO design changes could help?

Select 2 answers
A.Switch the storage engine from InnoDB to MyISAM
B.Use batch INSERT statements instead of single-row inserts
C.Upgrade to a larger RDS instance class
D.Remove unused or redundant indexes
E.Normalize the database schema to reduce data redundancy
AnswersB, D

Batch inserts reduce transaction overhead and log I/O.

Why this answer

Batch INSERT statements reduce the overhead of per-row index updates by combining multiple rows into a single transaction. This minimizes the number of index tree traversals and log flushes, directly lowering write amplification and improving insert throughput in InnoDB.

Exam trap

The trap here is that candidates often confuse scaling up (Option C) with optimizing write patterns, or assume that removing indexes (Option D) is the only way to reduce write amplification, when batch operations directly address the per-row overhead without sacrificing query performance.

1613
MCQmedium

A company runs a critical Amazon DynamoDB table with on-demand capacity. The table experiences throttling errors (ProvisionedThroughputExceededException) during peak hours. The table has a global secondary index (GSI) with a different partition key. What is the MOST likely cause of the throttling?

A.The base table's write capacity is insufficient for the workload.
B.The global secondary index is throttling due to uneven access patterns.
C.The partition key design causes hot partitions in the base table.
D.The table's read capacity is too low for the read-heavy workload.
AnswerB

GSIs on on-demand tables have their own capacity; hot partitions can cause throttling.

Why this answer

On-demand DynamoDB tables automatically scale capacity for the base table, but a GSI can throttle writes if its partition key creates hot partitions. The throttling on the base table occurs because the GSI write capacity is exceeded due to uneven access patterns. Options A, C, and D are incorrect because on-demand tables handle base table capacity scaling, hot partitions on the base table would cause throttling on the base table but the issue is with the GSI, and read capacity is not relevant given the write errors.

1614
MCQmedium

A company uses Amazon RDS for MySQL with Multi-AZ deployment. The database experiences intermittent write latency spikes. CloudWatch shows elevated 'WriteLatency' and 'WriteIOPS' but normal 'CPUUtilization'. Which is the MOST likely cause?

A.A parameter group change was applied without rebooting
B.The instance is exceeding the provisioned IOPS burst balance
C.A read replica is being used for write operations
D.Multi-AZ replication is causing synchronous writes to the standby
AnswerB

When EBS burst balance depletes, write latency spikes occur even with low CPU.

Why this answer

Intermittent write latency spikes with high WriteIOPS and normal CPU utilization indicate storage performance issues. For RDS MySQL using gp2 or io1 storage, exceeding the provisioned IOPS burst balance (for gp2) or provisioned IOPS (for io1) causes throttling and increased latency. Option A is incorrect because parameter group changes require a reboot and affect all operations, not just intermittent writes.

Option C is incorrect because read replicas are read-only and cannot be used for write operations. Option D is incorrect because Multi-AZ replication uses synchronous standby replication, which typically does not cause noticeable write latency spikes; latency from replication is usually minimal.

1615
Multi-Selecthard

A company uses Amazon DynamoDB to store order data. The table has a primary key (OrderID) and a Global Secondary Index (GSI) on CustomerID. The application often queries for all orders of a customer sorted by order date. The GSI projects only the keys. The queries are slow. What should the team do to improve query performance? (Choose two.)

Select 2 answers
A.Enable DynamoDB Accelerator (DAX) for the table
B.Increase the read capacity of the GSI
C.Modify the GSI to include OrderDate as a sort key
D.Use a Local Secondary Index (LSI) instead of a GSI
E.Change the GSI projection to include all attributes
AnswersC, E

Adding OrderDate as a sort key allows the GSI to return items sorted by order date without additional processing.

Why this answer

The queries are slow because the GSI projects only keys. When querying the GSI, DynamoDB must fetch the full items from the base table (a 'fetch' operation) for each key, which is inefficient. Option C (modifying the GSI to include OrderDate as a sort key) allows the GSI to sort results by order date natively, improving query performance.

Option E (changing the GSI projection to include all attributes) avoids the extra fetch by storing all attributes in the GSI, eliminating the need to access the base table. Option A (DAX) caches results but doesn't solve the sorting or projection issue. Option B (increasing read capacity) doesn't address the inefficiency of key-only projection.

Option D (using an LSI) is not possible because LSIs require the same partition key as the base table; CustomerID is different from OrderID, so an LSI cannot be used.

1616
MCQeasy

A database administrator notices that an Amazon RDS for Oracle DB instance's CPU utilization is consistently above 90% during peak hours. The application is read-heavy. Which action can reduce CPU load?

A.Disable Multi-AZ to free up resources
B.Increase the allocated storage
C.Enable Performance Insights to optimize queries
D.Create a read replica and direct read traffic to it
AnswerD

Offloading reads to a read replica reduces CPU load on the primary instance.

Why this answer

Creating a read replica offloads read traffic from the primary instance, reducing CPU utilization. For a read-heavy workload, this is an effective scaling strategy. Option A is incorrect because disabling Multi-AZ does not free up significant resources and reduces availability.

Option B (increase storage) does not directly reduce CPU load; it addresses storage capacity. Option C (Performance Insights) helps identify performance bottlenecks but does not reduce CPU load by itself.

1617
MCQeasy

A company needs to audit all SQL queries executed on an Amazon RDS for PostgreSQL instance. Which feature should the company enable?

A.pgAudit extension
B.Advanced Audit
C.Database Activity Streams
D.Fine-Grained Auditing
AnswerA

pgAudit provides detailed audit logs.

Why this answer

PgAudit (PostgreSQL Audit Extension) is the native, supported way to log all SQL queries on Amazon RDS for PostgreSQL. It provides detailed session and object audit logging via the shared_preload_libraries parameter, capturing every executed statement without requiring external tools or additional infrastructure.

Exam trap

The trap here is that candidates confuse Database Activity Streams (a real RDS feature for streaming activity to Kinesis) with a full SQL audit log, but Database Activity Streams does not capture every SQL query in a traditional audit log format and requires additional processing, whereas pgAudit directly writes detailed SQL logs to the RDS log files.

How to eliminate wrong answers

Option B (Advanced Audit) is wrong because it is not a real feature in Amazon RDS; it is a generic term that does not correspond to any specific RDS capability. Option C (Database Activity Streams) is wrong because it streams database activity to Amazon Kinesis for near-real-time monitoring and security analysis, but it does not capture every SQL query in the same granular, logged format as pgAudit; it is designed for threat detection, not full SQL audit trails. Option D (Fine-Grained Auditing) is wrong because it is an Oracle Database feature (FGA) that does not exist in PostgreSQL or Amazon RDS for PostgreSQL; it is a common misconception carried over from Oracle environments.

1618
MCQmedium

A company is migrating a 1 TB SQL Server database to Amazon RDS for SQL Server. The migration window is 4 hours. The on-premises network bandwidth is 1 Gbps. The database is not actively used during the migration window. The team wants to minimize migration time and cost. Which migration strategy is most appropriate?

A.Use SQL Server BCP utility to export data and then import using DTS.
B.Use AWS Snowball to transfer the database files.
C.Create a native backup, upload to S3, restore to RDS.
D.Use AWS DMS with full load only.
AnswerD

DMS can complete full load within 4 hours over 1 Gbps.

Why this answer

AWS DMS with full load only (Option D) is the most appropriate strategy for this migration. With a 1 Gbps network, the theoretical transfer time for 1 TB is about 2.2 hours (1 TB = 8000 Gb, 8000 Gb / 1 Gbps = 8000 seconds ≈ 2.22 hours), well within the 4-hour window. AWS DMS performs a direct, efficient full-load migration without requiring intermediate storage or additional tools.

Option A (BCP + DTS) is slower and more complex. Option B (Snowball) adds cost and logistics that are unnecessary given the network speed. Option C (native backup to S3 and restore) requires uploading the backup to S3 first, then restoring to RDS, which is typically slower than DMS's direct streaming approach.

1619
Multi-Selectmedium

A company is deploying an Amazon Aurora MySQL database with read replicas. They need to ensure high availability and disaster recovery across AWS Regions. Which THREE components should be included?

Select 3 answers
A.Aurora Replicas in the same Region
B.Aurora Global Database
C.Auto Scaling groups
D.Amazon Route 53 with health checks and failover policy
E.Amazon CloudFront
AnswersA, B, D

Aurora Replicas provide high availability within the region.

Why this answer

Aurora Replicas in the same Region provide read scaling and automatic failover within an AWS Region. They share the same underlying storage volume as the primary instance, so failover is fast (typically under 30 seconds) and does not require data rehydration. This is a core component for high availability within a single Region.

Exam trap

The trap here is that candidates often confuse Auto Scaling groups (which scale compute resources) with database replication or failover mechanisms, or they mistakenly think CloudFront can serve as a database-level disaster recovery solution when it only caches content at the edge.

1620
MCQeasy

A gaming company runs a leaderboard application on Amazon DynamoDB. The application experiences sudden spikes in read traffic during tournaments. The table uses on-demand capacity and the reads are eventually consistent. However, some users report stale data for several seconds. What is the most likely cause?

A.The application is using eventually consistent reads.
B.The table is using on-demand capacity instead of provisioned capacity.
C.The table has a global secondary index (GSI) that is not updated synchronously.
D.The read capacity units are insufficient for the traffic spikes.
AnswerA

Eventually consistent reads can return stale data within about 1 second.

Why this answer

Eventually consistent reads in DynamoDB can return stale data for up to one second under normal conditions, but during sudden spikes in read traffic, the replication lag can extend to several seconds. The application is using eventually consistent reads, which trade immediate consistency for higher throughput and lower latency, making stale data more likely during high-traffic periods like tournaments.

Exam trap

The trap here is that candidates may confuse eventual consistency with capacity issues, but DynamoDB's on-demand mode eliminates throttling, so stale data points directly to the consistency model rather than resource constraints.

How to eliminate wrong answers

Option B is wrong because on-demand capacity automatically scales to handle traffic spikes without throttling, so it does not cause stale data. Option C is wrong because global secondary indexes (GSIs) are updated synchronously with the base table in DynamoDB, meaning they always reflect the latest write; stale data from a GSI would only occur if the application used eventually consistent reads on the GSI itself. Option D is wrong because read capacity units are not applicable to on-demand capacity mode, which has no fixed capacity limits; insufficient capacity would cause throttling errors (e.g., ProvisionedThroughputExceededException), not stale data.

1621
Multi-Selecthard

Which THREE of the following are best practices for securing an Amazon Aurora MySQL database? (Select THREE.)

Select 3 answers
A.Enable encryption at rest using AWS KMS.
B.Enable audit logging to track database activity.
C.Store database credentials in application code for simplicity.
D.Use IAM database authentication for access.
E.Disable automated backups to reduce attack surface.
AnswersA, B, D

Encryption at rest protects data if storage is compromised.

Why this answer

Options A, B, and D are correct. Enabling encryption at rest with AWS KMS (A) protects data stored on disk. Audit logging (B) tracks database activity for security monitoring.

IAM database authentication (D) provides a secure, managed method for access control without storing credentials. Option C is wrong because storing credentials in application code is insecure and violates best practices. Option E is wrong because disabling automated backups reduces data durability and recovery capabilities.

1622
MCQmedium

Refer to the exhibit. A database engineer runs the command and receives the above output. The engineer wants to convert the database to Multi-AZ. Which command should be used?

A.aws rds reboot-db-instance --db-instance-identifier mydb --force-failover
B.aws rds create-db-instance --db-instance-identifier mydb --multi-az
C.aws rds modify-db-instance --db-instance-identifier mydb --multi-az
D.aws rds create-db-cluster --db-instance-identifier mydb --multi-az
AnswerC

'aws rds modify-db-instance' with the '--multi-az' flag modifies an existing DB instance to use Multi-AZ deployment.

Why this answer

'aws rds modify-db-instance' with the '--multi-az' flag modifies an existing DB instance to use Multi-AZ deployment. Option A is incorrect because 'reboot-db-instance' with '--force-failover' triggers a reboot and failover but does not change the Multi-AZ setting itself. Option B is incorrect because 'create-db-instance' is used to create a new instance, not modify an existing one.

Option D is incorrect because 'create-db-cluster' is for creating an Aurora DB cluster, not for converting a single DB instance to Multi-AZ.

1623
MCQeasy

A company has an Amazon DynamoDB table that stores session data for a web application. The table's read capacity units (RCUs) are consistently near 100% utilization during peak hours, causing throttling. The application can tolerate eventually consistent reads. Which action should the application team take to reduce throttling?

A.Decrease the provisioned RCUs for the table.
B.Enable DynamoDB Accelerator (DAX) to cache the session data.
C.Change the read consistency to eventually consistent.
D.Increase the provisioned RCUs for the table.
AnswerC

Using eventually consistent reads halves RCU consumption, doubling effective read capacity and reducing throttling.

Why this answer

Eventually consistent reads consume only half the Read Capacity Units (RCUs) compared to strongly consistent reads. Switching to eventually consistent reads effectively doubles the read capacity, reducing throttling without additional cost. Option A is incorrect because decreasing provisioned RCUs would reduce capacity and worsen throttling.

Option B is incorrect because while DAX can cache session data and reduce read load on the table, it does not directly address the RCU consumption for reads that still hit the table; moreover, implementing DAX adds complexity and cost. Option D is incorrect because increasing RCUs would solve throttling but at a higher cost, whereas eventually consistent reads provide a cost-effective solution.

1624
MCQmedium

A company uses Amazon RDS for Oracle with a Multi-AZ deployment for a critical OLTP application. During a recent failover test, they noticed that the application experienced a two-minute downtime. The team wants to reduce downtime to under 30 seconds during automatic failovers. What should they do?

A.Add a read replica to offload reads
B.Reduce the DNS TTL value to 5 seconds
C.Enable Automatic Failover in the RDS console
D.Migrate to Amazon Aurora with Multi-AZ and use the Aurora auto-failover feature
AnswerD

Aurora failover is typically under 30 seconds, and it provides faster recovery than RDS Multi-AZ.

Why this answer

Amazon Aurora with Multi-AZ provides faster failover than RDS for Oracle because Aurora uses a shared storage architecture and a cluster endpoint that automatically redirects traffic to the replica within 30 seconds, often in as little as 15 seconds. In contrast, RDS for Oracle Multi-AZ relies on DNS record updates and a standby instance that must be promoted, which typically takes 60–120 seconds. Migrating to Aurora eliminates the DNS propagation delay and the need for storage failover, meeting the sub-30-second requirement.

Exam trap

The trap here is that candidates assume reducing DNS TTL (Option B) will solve the problem, but they overlook that the primary bottleneck in RDS for Oracle failover is the database promotion and recovery time, not just DNS caching.

How to eliminate wrong answers

Option A is wrong because adding a read replica offloads read traffic but does not reduce failover time; failover still occurs on the primary instance and requires the same DNS and promotion steps. Option B is wrong because reducing DNS TTL to 5 seconds only minimizes client-side caching delay, but the actual failover process in RDS for Oracle (including storage and instance promotion) still takes 60–120 seconds, so the total downtime remains well over 30 seconds. Option C is wrong because 'Automatic Failover' is already enabled by default in a Multi-AZ deployment; there is no separate toggle to enable it, and the two-minute downtime is inherent to RDS for Oracle's failover mechanism, not a configuration issue.

1625
MCQhard

A gaming company uses Amazon ElastiCache for Redis as a leaderboard for real-time game scores. The leaderboard is updated frequently by millions of users. The application uses sorted sets with player scores. Recently, the leaderboard update latency increased and the cache evictions spiked. The company needs to ensure low-latency updates and high availability. The current setup is a single Redis node. Which design should be implemented?

A.Upgrade to a larger single Redis node instance type to handle the load.
B.Replace Redis with DynamoDB for the leaderboard, using a global secondary index on score.
C.Use a Redis Cluster with multiple shards. Enable AOF persistence and use a read replica for the leaderboard queries.
D.Use ElastiCache for Redis with cluster mode disabled and enable Multi-AZ.
AnswerC

Redis Cluster distributes data across shards, reducing load per node. Read replicas can handle queries, and AOF ensures durability.

Why this answer

Redis Cluster with multiple shards distributes the write load across shards, reducing per-node pressure and evictions. Enabling AOF persistence ensures durability, while using a read replica for leaderboard queries offloads read traffic from the primary shard, maintaining low-latency updates. This design provides both horizontal scaling and high availability, addressing the increased update latency and eviction spikes.

Exam trap

The trap here is that candidates may assume Multi-AZ (Option D) alone solves high availability and performance, but without sharding (cluster mode enabled), a single node remains a bottleneck for write-heavy workloads, and evictions will continue.

How to eliminate wrong answers

Option A is wrong because upgrading to a larger single Redis node instance type only provides vertical scaling, which has a hard ceiling and does not eliminate the single point of failure or the risk of evictions under sustained high write throughput. Option B is wrong because DynamoDB with a global secondary index on score is not optimized for real-time sorted set operations like ZADD and ZRANGE; it lacks the atomic, in-memory sorted set semantics that Redis provides for leaderboards, leading to higher latency and complexity for frequent updates. Option D is wrong because ElastiCache for Redis with cluster mode disabled and Multi-AZ only provides failover redundancy but does not shard data; a single node still handles all writes, so evictions and latency will persist under high load.

1626
Multi-Selectmedium

A company is deploying a new MySQL database on Amazon RDS. The database must be highly available and support cross-Region disaster recovery. Which TWO actions should be taken? (Choose two.)

Select 2 answers
A.Enable encryption at rest using AWS KMS.
B.Deploy a Multi-AZ instance in multiple regions.
C.Enable automated backups with a retention period of 35 days.
D.Create a cross-region read replica.
E.Enable Multi-AZ for automatic failover within a region.
AnswersD, E

Provides disaster recovery in another region.

Why this answer

A cross-region read replica provides a secondary copy of the database in a different AWS Region, which can be promoted to a standalone primary in a disaster scenario, enabling cross-Region disaster recovery. Option E is correct because Multi-AZ deployment provides automatic failover within a single region by synchronously replicating data to a standby instance in a different Availability Zone, ensuring high availability. Together, these two actions satisfy both high availability (within-region failover) and cross-Region DR requirements.

Exam trap

The trap here is that candidates often confuse Multi-AZ with cross-region replication, thinking that Multi-AZ can be deployed across regions (Option B), or they overlook that automated backups (Option C) are region-bound and do not provide automatic failover or cross-Region DR.

1627
MCQmedium

A company is using Amazon ElastiCache for Redis to cache database query results. They notice that the cache hit ratio is low and many requests are hitting the database. The 'Evictions' metric is high. What is the MOST likely cause?

A.The replication factor is insufficient.
B.Encryption in transit is enabled.
C.The application is experiencing high network latency.
D.The cache node is too small for the workload.
AnswerD

A small cache leads to frequent evictions, reducing the cache hit ratio.

Why this answer

The high 'Evictions' metric indicates that the cache node is constantly full, causing it to evict old data to make room for new data. This leads to a low cache hit ratio because frequently accessed items are evicted and must be fetched from the database. Option D correctly identifies that the cache node is too small for the workload.

Option A is incorrect because insufficient replication factor would affect availability, not evictions. Option B is incorrect because encryption in transit does not impact cache capacity or evictions. Option C is incorrect because high network latency would affect request latency but not directly cause evictions.

1628
MCQhard

A company is designing a social media application that requires storing user relationships (follows) and making graph queries like 'mutual friends.' Which database is most suitable?

A.Amazon ElastiCache for Redis
B.Amazon DynamoDB
C.Amazon Neptune
D.Amazon RDS for MySQL
AnswerC

Neptune is a graph database optimized for highly connected data and graph queries.

Why this answer

Amazon Neptune is a fully managed graph database service optimized for storing and querying highly connected data. It supports both property graph and RDF models, and it uses Gremlin or SPARQL to efficiently traverse relationships like 'mutual friends' in a social media application, making it the ideal choice for graph queries.

Exam trap

The trap here is that candidates often choose DynamoDB for its scalability and low latency, overlooking that graph queries like 'mutual friends' require native graph traversal capabilities that DynamoDB's key-value model cannot efficiently provide.

How to eliminate wrong answers

Option A is wrong because Amazon ElastiCache for Redis is an in-memory key-value store and cache, not a graph database; it lacks native graph traversal capabilities and would require complex application-side logic to compute mutual friends. Option B is wrong because Amazon DynamoDB is a NoSQL key-value and document database that does not support graph queries; it would require multiple queries and client-side joins to resolve relationships, leading to poor performance and scalability for graph workloads. Option D is wrong because Amazon RDS for MySQL is a relational database that uses SQL joins to model relationships, which becomes inefficient and unscalable for deep graph traversals like mutual friends due to the exponential number of join operations required.

1629
MCQhard

A company runs a multi-tenant SaaS application on Amazon DynamoDB. Each tenant's data is stored in a separate table named with a tenant-specific prefix (e.g., tenant1_orders, tenant2_orders). The application uses DynamoDB Streams to replicate data to a central analytics table. Recently, the company added a new large tenant that generates 10x more write traffic than any other tenant. The DynamoDB Streams for the large tenant's table is falling behind by several hours, causing stale data in the analytics table. The company has already increased the write capacity of the large tenant's table to 50,000 WCUs, but the streams lag persists. The analytics table is also in DynamoDB and uses a Global Secondary Index (GSI) for querying. The streams processing Lambda function performs simple transformations and writes to the analytics table. The Lambda function is not throttled. Which action would resolve the streams lag?

A.Enable DynamoDB on-demand mode for the large tenant's table to allow automatic scaling of stream shards.
B.Remove the GSI from the analytics table to reduce write amplification.
C.Increase the Lambda function's reserved concurrency to the maximum.
D.Increase the write capacity of the large tenant's table to 100,000 WCUs.
AnswerA

On-demand mode adjusts the number of stream shards based on write traffic, which can help with lag.

Why this answer

DynamoDB Streams shards are directly tied to the physical partitions of the table. When a table is in provisioned mode, the number of stream shards is fixed and determined by the table's partitions, which cannot scale independently. Enabling on-demand mode allows DynamoDB to automatically split partitions and thus increase the number of stream shards, enabling higher stream throughput to keep up with the large tenant's write volume.

This directly addresses the root cause of the streams lag without requiring manual partition management.

Exam trap

The trap here is that candidates assume increasing write capacity alone will resolve stream lag, but they overlook that stream shard count is tied to physical partitions, which only increase with on-demand mode or by triggering partition splits through sustained high throughput.

How to eliminate wrong answers

Option B is wrong because removing the GSI from the analytics table would reduce write amplification for writes to the analytics table, but the bottleneck is the DynamoDB Streams processing of the large tenant's source table, not the write capacity of the analytics table. Option C is wrong because the Lambda function is not throttled, so increasing reserved concurrency will not help; the issue is that the stream shards cannot process records fast enough due to insufficient shard count. Option D is wrong because increasing write capacity to 100,000 WCUs does not increase the number of stream shards; stream shard count is determined by the number of physical partitions, which only changes when partitions split, and provisioned WCUs alone do not trigger partition splits beyond the initial allocation.

1630
MCQmedium

A company is building a real-time chat application that requires storing messages with a maximum of 10,000 characters per message. The application needs sub-millisecond latency for reads and writes. The data must be durable and replicated across three Availability Zones. The development team wants to minimize operational overhead. Which AWS database service is most appropriate?

A.Amazon ElastiCache for Redis with replication
B.Amazon DynamoDB with DAX
C.Amazon RDS for PostgreSQL with Multi-AZ
D.Amazon Aurora MySQL with Multi-AZ
AnswerB

Serverless, sub-millisecond latency, durable, multi-AZ.

Why this answer

Amazon DynamoDB with DAX is the most appropriate choice because it provides single-digit millisecond latency for reads and writes, supports up to 400 KB per item (easily accommodating 10,000 characters), and offers built-in replication across three Availability Zones for durability. DAX (DynamoDB Accelerator) further reduces read latency to sub-millisecond by serving as an in-memory cache, while DynamoDB itself handles write durability and replication automatically, minimizing operational overhead.

Exam trap

The trap here is that candidates often choose ElastiCache for Redis (Option A) because of its sub-millisecond latency, overlooking the requirement for durable, multi-AZ replicated storage that Redis alone does not provide natively without additional configuration and operational overhead.

How to eliminate wrong answers

Option A is wrong because Amazon ElastiCache for Redis is an in-memory data store that does not provide durable storage by default; while it can be configured with replication, it lacks the native multi-AZ durability guarantees required for persistent message storage and would require additional infrastructure for data persistence. Option C is wrong because Amazon RDS for PostgreSQL with Multi-AZ provides high availability but cannot achieve sub-millisecond latency for both reads and writes due to disk-based storage and synchronous replication overhead, and it requires manual scaling and management. Option D is wrong because Amazon Aurora MySQL with Multi-AZ offers better performance than standard RDS but still cannot guarantee sub-millisecond latency for writes due to its distributed storage architecture and replication across three AZs, and it introduces more operational complexity than a fully managed NoSQL solution like DynamoDB.

1631
MCQeasy

An administrator notices that the CloudWatch metric 'ReadLatency' for an Amazon RDS for SQL Server instance has increased significantly. Which of the following is the most likely cause?

A.The DB instance is experiencing high CPU utilization.
B.The DB instance is running out of memory.
C.The DB instance is using a burstable instance class that has exhausted its credits.
D.The DB instance does not have enough provisioned IOPS.
AnswerD

Insufficient provisioned IOPS means the database cannot service I/O requests as fast as needed, causing queuing and increased read latency. This is the most likely cause.

Why this answer

High ReadLatency indicates that read operations are taking longer than expected. This is most directly caused by insufficient I/O throughput, i.e., not enough provisioned IOPS for the workload. When the database requests more IOPS than are provisioned, I/O operations queue up, increasing latency.

Option D is correct because insufficient provisioned IOPS is a primary cause of increased ReadLatency. Option A is incorrect: high CPU utilization can affect overall performance but does not directly cause read latency unless it leads to I/O contention. Option B is incorrect: running out of memory can cause disk swapping, which may increase latency, but it is less direct than I/O throttling.

Option C is incorrect: burstable instance classes (like T-series) use CPU credits; exhaustion affects CPU performance, not storage I/O latency.

1632
Multi-Selectmedium

A company is deploying a new application using Amazon RDS for PostgreSQL. They require high availability with automatic failover and the ability to offload read traffic. Which TWO features should they enable? (Choose two.)

Select 2 answers
A.Multi-AZ deployment
B.Read replicas
C.Performance Insights
D.Global Database
E.Auto Scaling for read replicas
AnswersA, B

Multi-AZ provides automatic failover to a standby in another Availability Zone.

Why this answer

Multi-AZ deployment (Option A) provides high availability by automatically provisioning and maintaining a synchronous standby replica in a different Availability Zone. If the primary DB instance fails, Amazon RDS automatically fails over to the standby, ensuring minimal downtime. Read replicas (Option B) offload read traffic by creating asynchronous copies of the database that can serve read queries, reducing load on the primary instance.

Exam trap

The trap here is that candidates may confuse Multi-AZ with read replicas, thinking they serve the same purpose, or incorrectly assume that Global Database or Auto Scaling for read replicas are available for standard RDS PostgreSQL, when they are either Aurora-specific or not supported.

1633
MCQmedium

A developer is troubleshooting an application that uses Amazon DynamoDB. The application sometimes receives ProvisionedThroughputExceededException errors. The table has on-demand capacity mode. The errors occur in short bursts. What is the most likely cause?

A.The table has a low read/write capacity mode limit that needs to be increased.
B.The global secondary index (GSI) has a different throughput limit.
C.The table has reached the maximum provisioned throughput.
D.The request rate exceeds the partition's throughput capacity in a short burst.
AnswerD

On-demand can throttle if a single partition's throughput is exceeded.

Why this answer

On-demand capacity mode can handle up to the table's previous peak traffic. However, if traffic spikes suddenly, DynamoDB might throttle. Option D is correct because on-demand has a limit on the maximum throughput per partition.

Option A is wrong because indexes share the table's capacity. Option B is wrong because on-demand does not have provisioned limits. Option C is wrong because the table is on-demand, not provisioned.

1634
MCQhard

Refer to the exhibit. An application on an EC2 instance is trying to read an object from the S3 bucket 'data-lake-prod'. The instance is in a VPC with an IP address of 10.0.1.5. The application receives an Access Denied error. What is the cause?

A.An SCP is denying S3 access to the account.
B.A Deny statement with a source IP condition is blocking access.
C.The S3 bucket policy requires a specific VPC endpoint.
D.The IAM role does not have permission to read from the bucket.
AnswerB

The Deny with condition matches the IP and overrides the Allow.

Why this answer

The Deny statement applies to the source IP 10.0.1.5 (within 10.0.0.0/8), overriding the Allow. Option A is wrong because the IAM role does allow s3:GetObject on the bucket. Option C is wrong because there is no explicit condition on the Allow statement.

Option D is wrong because the policy is attached to the role, not an SCP.

1635
MCQmedium

Refer to the exhibit. A database administrator is writing an IAM policy to manage Amazon RDS snapshots. The policy allows actions on all RDS resources but denies deletion of any DB instance with the prefix 'production-'. A developer attempts to delete a snapshot named 'production-snapshot-2020'. Will this action be allowed?

A.Yes, but only if the developer has the necessary permissions.
B.No, because the Deny statement overrides the Allow statement.
C.Yes, because the Deny statement only applies to deleting DB instances, not snapshots.
D.No, because the snapshot name starts with 'production-', which matches the resource pattern in the Deny statement.
AnswerC

The Deny is specific to 'rds:DeleteDBInstance', so snapshot deletion is allowed.

Why this answer

The Deny statement in the policy only applies to the 'rds:DeleteDBInstance' action, not to 'rds:DeleteDBSnapshot'. The Allow statement explicitly allows all actions including 'rds:DeleteDBSnapshot' on all resources. Since there is no explicit deny for deleting snapshots, the allow takes effect.

Option A is incorrect because the developer's permissions are not relevant; the policy allows the action. Option B is incorrect because the Deny does not apply to snapshots. Option D is incorrect because the Deny condition only matches deletions of DB instances, not snapshots.

1636
MCQeasy

A company is deploying a new application that requires a MySQL-compatible database with automatic scaling of compute and storage resources. Which AWS service should be used?

A.Amazon Aurora MySQL
B.Amazon RDS for MySQL
C.Amazon DynamoDB
D.Amazon Redshift
AnswerA

Supports automatic scaling for compute and storage.

Why this answer

Amazon Aurora MySQL is the correct choice because it is a MySQL-compatible relational database that automatically scales compute (up to 128 vCPUs and 244 GiB of memory) and storage (up to 128 TiB in 10 GiB increments) without downtime. Aurora's storage subsystem is distributed and self-healing, scaling as data grows, and its compute can be scaled via Auto Scaling policies or by modifying the instance class with minimal disruption.

Exam trap

The trap here is that candidates often confuse Amazon RDS for MySQL with Aurora MySQL, assuming both offer automatic scaling, but RDS requires manual scaling of compute and storage, while Aurora provides true automatic scaling for both.

How to eliminate wrong answers

Option B is wrong because Amazon RDS for MySQL does not support automatic scaling of compute resources; you must manually modify the DB instance class and storage, which can cause downtime. Option C is wrong because Amazon DynamoDB is a NoSQL key-value and document database, not MySQL-compatible, and it does not support SQL queries or relational features like joins and transactions. Option D is wrong because Amazon Redshift is a petabyte-scale data warehouse optimized for analytical queries, not a MySQL-compatible transactional database, and it does not provide automatic scaling of compute or storage in the same manner as Aurora.

1637
MCQeasy

A company wants to ensure that only specific IAM users can perform certain operations on an Amazon RDS DB instance, such as creating snapshots or modifying the instance. Which AWS feature should be used to define these permissions?

A.VPC security groups
B.IAM policies
C.DB parameter groups
D.DB subnet groups
AnswerB

IAM policies define permissions for AWS actions on resources like RDS.

Why this answer

IAM policies are used to grant or deny permissions to AWS resources, including Amazon RDS DB instances. By attaching an IAM policy to a user, group, or role, you can control which actions (e.g., CreateDBSnapshot, ModifyDBInstance) are allowed. Options A, C, and D are incorrect: VPC security groups control network traffic, DB parameter groups manage database engine settings, and DB subnet groups define which subnets the DB instance can use—none of these define permissions for specific operations.

1638
MCQmedium

A company is running a production Amazon RDS for MySQL DB instance. The application team reports intermittent high latency and connection timeouts. A quick check shows that the DB instance's CPU utilization is consistently above 90% during peak hours. The database size is 500 GB and the instance class is db.r5.large. Which combination of actions should a database specialist take to resolve the performance issue?

A.Increase the allocated storage to 1 TB and enable auto-scaling for storage.
B.Scale up the DB instance to db.r5.xlarge and review slow query logs to optimize poorly performing queries.
C.Enable Multi-AZ and increase the allocated storage to 1 TB to improve I/O performance.
D.Enable Performance Insights and create a CloudWatch alarm to notify when CPU exceeds 80%.
AnswerB

Scaling up provides more CPU and memory; slow query logs help identify and fix inefficient queries.

Why this answer

Scaling up the DB instance to db.r5.xlarge provides more compute capacity, directly addressing high CPU utilization. Additionally, reviewing slow query logs helps identify and optimize inefficient queries that may be causing CPU spikes. Option A is incorrect because increasing storage does not improve CPU performance; it only addresses storage capacity or I/O issues.

Option C is incorrect because enabling Multi-AZ provides high availability but does not increase compute capacity, and increasing storage does not help with CPU. Option D is incorrect because Performance Insights helps diagnose performance issues but does not resolve them, and CloudWatch alarms only provide notifications without fixing the underlying problem.

1639
Matchingmedium

Match each AWS database-related CLI command to its function.

Drag a concept onto its matching description — or click a concept then click the description.

Concepts
Matches

Creates a new RDS DB instance

Inserts or replaces an item in a DynamoDB table

Returns details about Redshift clusters

Creates an ElastiCache cache cluster

Lists manual and automated DB snapshots

Why these pairings

The correct matches pair each AWS CLI command with its corresponding database service. Common confusions involve swapping commands between RDS, DynamoDB, Redshift, and ElastiCache. Remember that each service has its own command namespace (e.g., 'aws rds', 'aws dynamodb', 'aws redshift', 'aws elasticache').

1640
MCQeasy

A company is running an Amazon Aurora MySQL-compatible edition DB cluster. The application reports that a recent schema change caused a significant increase in the number of deadlocks. Which parameter should the DBA check and possibly adjust to reduce deadlocks?

A.innodb_lock_wait_timeout
B.autocommit
C.max_connections
D.transaction_isolation
AnswerD

transaction_isolation sets the isolation level for transactions. Lowering the isolation level (e.g., from REPEATABLE READ to READ COMMITTED) reduces the use of gap locks and other locking mechanisms, thereby decreasing the likelihood of deadlocks.

Why this answer

Adjusting the transaction isolation level can reduce deadlocks by decreasing the locking overhead. For example, using READ COMMITTED instead of REPEATABLE READ reduces the use of gap locks, which are a common source of deadlocks. Option A, innodb_lock_wait_timeout, determines how long a transaction waits for a lock before timing out; it does not prevent deadlocks but rather handles lock waits.

Option B, autocommit, affects when transactions are committed but does not directly impact deadlock occurrence. Option C, max_connections, limits concurrent connections, which can reduce overall contention but is not a direct deadlock reduction parameter.

1641
Multi-Selecteasy

Which TWO AWS services can be used to migrate a MongoDB database to Amazon DocumentDB? (Choose 2.)

Select 2 answers
A.AWS DataSync
B.AWS Database Migration Service (AWS DMS)
C.Amazon S3
D.MongoDB native tools (mongodump and mongorestore)
E.Amazon Kinesis Data Streams
AnswersB, D

DMS supports MongoDB to DocumentDB migration.

Why this answer

AWS Database Migration Service (AWS DMS) supports MongoDB as a source and Amazon DocumentDB as a target, enabling continuous replication with change data capture (CDC) for minimal downtime. MongoDB native tools (mongodump and mongorestore) are also valid for offline, one-time migrations by exporting the database as BSON files and importing them directly into DocumentDB.

Exam trap

The trap here is that candidates often confuse AWS DataSync or S3 as viable migration tools for NoSQL databases, but neither supports the native MongoDB protocol or BSON format required for DocumentDB ingestion.

1642
MCQhard

A company is running a MongoDB-compatible Amazon DocumentDB cluster with one writer and two readers. The application writes a large amount of data during batch processing, and after a batch completes, the writer's CPU is high, and the readers have significant replica lag. The team wants to reduce replica lag without affecting the batch performance. What should they do?

A.Change the storage type to Provisioned IOPS on all instances
B.Increase the instance size of the readers to improve apply throughput
C.Reduce the batch size to lower the write rate
D.Increase the instance size of the writer to handle the batch faster
AnswerB

Larger readers can apply oplog entries faster, reducing lag.

Why this answer

Replica lag in Amazon DocumentDB occurs when readers are unable to apply writes from the writer's oplog quickly enough. Increasing the instance size of the readers provides more CPU and memory resources, improving their apply throughput and reducing lag without affecting the writer's batch performance. Option A is incorrect because changing the storage type to Provisioned IOPS may improve I/O but does not directly address the reader's apply capacity.

Option C is incorrect because reducing batch size would lower the write rate and degrade batch performance, which the team wants to avoid. Option D is incorrect because increasing the writer's instance size may help the writer handle more writes, but it does not help readers apply those writes faster; it could even increase the write rate and worsen replica lag.

1643
MCQmedium

A company is using Amazon DynamoDB with global tables. The company wants to monitor replication latency between regions. Which CloudWatch metric should the company use?

A.GlobalTableReplicationLag
B.ReplicaLag
C.WriteThrottleEvents
D.ReplicationLatency
AnswerD

This metric tracks replication lag.

Why this answer

The correct metric is ReplicationLatency, which measures the time elapsed between an update in the source region and its appearance in the replica region for DynamoDB global tables. This metric directly reflects the replication lag between regions, making it the appropriate choice for monitoring cross-region replication performance.

Exam trap

The trap here is confusing DynamoDB's ReplicationLatency with RDS's ReplicaLag, as both involve cross-region replication but apply to entirely different services and metrics.

How to eliminate wrong answers

Option A is wrong because GlobalTableReplicationLag is not a valid CloudWatch metric for DynamoDB; AWS uses ReplicationLatency for this purpose. Option B is wrong because ReplicaLag is a metric for Amazon RDS read replicas, not for DynamoDB global tables. Option C is wrong because WriteThrottleEvents tracks throttled write requests, not replication latency between regions.

1644
MCQeasy

A company is migrating an on-premises MySQL database to Amazon RDS for MySQL. They need to validate that the schema is compatible. Which AWS service should they use?

A.AWS Schema Conversion Tool (SCT)
B.Amazon Inspector
C.AWS Database Migration Service (DMS)
D.AWS Trusted Advisor
AnswerA

SCT checks and converts schemas.

Why this answer

The AWS Schema Conversion Tool (SCT) is designed specifically to assess and convert database schemas from one engine to another, including on-premises MySQL to Amazon RDS for MySQL. It analyzes the source schema for compatibility issues, identifies unsupported features or deprecated syntax, and generates a detailed assessment report. This makes it the correct choice for validating schema compatibility before migration.

Exam trap

The trap here is that candidates often confuse AWS DMS with schema validation, assuming DMS handles both data migration and schema conversion, but DMS only moves data and requires a pre-validated schema; SCT is the dedicated tool for schema compatibility analysis.

How to eliminate wrong answers

Option B is wrong because Amazon Inspector is a vulnerability management service that scans workloads for software vulnerabilities and unintended network exposure; it does not perform database schema analysis or compatibility checks. Option C is wrong because AWS Database Migration Service (DMS) is used for migrating data (rows and tables) from a source to a target database, but it does not validate schema compatibility or convert schema objects; schema conversion is handled separately by SCT. Option D is wrong because AWS Trusted Advisor provides best-practice recommendations for cost optimization, performance, security, and fault tolerance, but it does not analyze database schema compatibility or migration readiness.

1645
MCQeasy

A company is deploying a new microservices application on Amazon ECS. Each microservice needs its own database. Which AWS database service should be used to minimize operational overhead?

A.Amazon DynamoDB
B.Amazon Aurora
C.Amazon Redshift
D.Amazon RDS for MySQL
AnswerA

DynamoDB is fully managed, serverless, and suitable for microservices.

Why this answer

Amazon DynamoDB is a fully managed NoSQL database that provides single-digit millisecond latency at any scale, making it ideal for microservices that each require their own isolated database. It eliminates operational overhead by handling hardware provisioning, setup, configuration, replication, and patching automatically, with no servers to manage. DynamoDB also supports on-demand capacity mode, which scales automatically based on traffic, perfectly aligning with the dynamic nature of microservices on Amazon ECS.

Exam trap

The trap here is that candidates often assume a relational database (like Aurora or RDS) is always the best choice for microservices, overlooking the operational simplicity and scalability of a fully managed NoSQL service like DynamoDB, which is specifically designed for high-traffic, schema-less, and serverless architectures.

How to eliminate wrong answers

Option B (Amazon Aurora) is wrong because while it is fully managed, it is a relational database that requires schema definition and connection management, adding operational overhead compared to a schema-less NoSQL solution like DynamoDB. Option C (Amazon Redshift) is wrong because it is a petabyte-scale data warehouse designed for analytical workloads, not for transactional microservice databases, and it introduces significant operational complexity with cluster management. Option D (Amazon RDS for MySQL) is wrong because although it reduces some operational tasks, it still requires manual scaling, connection pooling, and schema management, and does not provide the same level of automatic scaling and serverless operation as DynamoDB.

1646
MCQhard

A company is migrating a 5 TB Oracle database to Amazon Aurora MySQL using AWS DMS. The full load completes successfully, but the CDC phase is slow and the target Aurora cluster is consuming high CPU. The DMS instance type is dms.c4.2xlarge. Which change will most likely improve CDC performance?

A.Increase the Aurora cluster's max_connections parameter
B.Increase the number of parallel apply threads in the DMS task
C.Increase the DMS task's target storage allocation
D.Change the Aurora cluster to a larger instance class
AnswerB

More parallel apply threads can increase the rate of applying changes to Aurora, reducing CPU usage per thread.

Why this answer

Increasing the number of parallel apply threads in the DMS task allows the CDC phase to apply multiple transactions concurrently to the target Aurora cluster, reducing the backlog of changes. The high CPU on the target suggests the apply process is the bottleneck, not the source or network, and DMS's parallel apply threads directly address this by enabling multi-threaded writes, which improves throughput without requiring larger instances.

Exam trap

The trap here is that candidates often assume high CPU on the target means the instance is undersized and needs a larger class (Option D), but the real issue is the DMS apply process being single-threaded, which causes CPU to spike due to inefficient serial writes rather than insufficient compute capacity.

How to eliminate wrong answers

Option A is wrong because increasing max_connections does not improve CDC apply performance; it only allows more client connections to the Aurora cluster, but the bottleneck is the single-threaded apply of changes, not connection limits. Option C is wrong because target storage allocation affects disk space, not the rate at which DMS applies transactions; CDC performance is CPU and I/O bound, not storage capacity bound. Option D is wrong because while a larger Aurora instance class might reduce CPU pressure, the root cause is the DMS task's single-threaded apply; upgrading the cluster without enabling parallel apply threads would still leave the apply process as a bottleneck, and it is more cost-effective to tune DMS parallelism first.

1647
MCQeasy

A database administrator needs to monitor the number of connections to an Amazon RDS for MySQL DB instance. Which Amazon CloudWatch metric should be used?

A.NetworkThroughput
B.DatabaseConnections
C.ConnectionAttempts
D.ActiveTransactions
AnswerB

DatabaseConnections is the correct metric for monitoring the number of connections.

Why this answer

The DatabaseConnections metric in CloudWatch shows the number of current connections to the DB instance.

1648
MCQhard

Refer to the exhibit. A database administrator has this IAM policy attached to their user. They are trying to authorize ingress to a DB security group named 'my-security-group' in the us-east-1 region. The operation fails. What is the most likely reason?

A.The user is not allowed to create DB security groups, which is a prerequisite.
B.The policy does not grant 'rds:AuthorizeDBSecurityGroupIngress' on the specific security group.
C.The security group ARN in the policy is incorrect; it should include the security group ID instead of the name.
D.The user is missing the 'rds:DescribeDBSecurityGroups' permission, which is required to perform the authorize operation.
AnswerD

IAM requires the corresponding Describe action for many APIs.

Why this answer

The policy grants the `rds:AuthorizeDBSecurityGroupIngress` action on a specific DB security group. However, to authorize ingress, the user must also be able to describe the security group to verify its existence and details. The policy does not include `rds:DescribeDBSecurityGroups`, which is a prerequisite for the authorization operation.

Therefore, the operation fails due to missing describe permission.

1649
MCQmedium

The exhibit shows the output of an AWS CLI command for an Amazon RDS DB instance. Based on the output, what is the current configuration of the DB instance?

A.It is an Amazon Aurora MySQL-Compatible DB cluster.
B.It is a MySQL DB instance that is currently in a failed state.
C.It is a Multi-AZ DB instance with a standby in a different AZ.
D.It is a Single-AZ MySQL DB instance.
AnswerD

MultiAZ is false and SecondaryAvailabilityZone is null, confirming Single-AZ.

Why this answer

The output shows engine=mysql, MultiAZ=false, SecondaryAvailabilityZone=null, and DBInstanceStatus=available. This indicates a Single-AZ MySQL DB instance, not a Multi-AZ deployment, not Aurora, and not in a failed state. Therefore, option D is correct.

1650
MCQmedium

A company uses Amazon DynamoDB for a real-time analytics application. The table is provisioned with 1000 RCU and 500 WCU. Recently, the application experienced throttling on reads during a marketing campaign. The 'ConsumedReadCapacityUnits' metric shows spikes up to 2000 RCU. The 'ReadThrottleEvents' metric is high. The table has a global secondary index (GSI) that is also throttled. The application uses eventually consistent reads. The team wants to minimize costs while preventing throttling. The campaign traffic is predictable and occurs for 2 hours each day. What should the team do?

A.Increase the provisioned RCU to 2000 permanently.
B.Implement DynamoDB Accelerator (DAX) to cache reads.
C.Enable auto scaling with a scheduled scaling action to increase RCU during the campaign.
D.Switch the table to on-demand capacity mode.
AnswerC

Auto scaling adjusts capacity automatically, and scheduled scaling ensures capacity is ready for the campaign, minimizing cost.

Why this answer

The traffic is predictable and occurs for 2 hours daily. Scheduled auto scaling can increase RCU during the campaign and reduce it afterward, minimizing cost while preventing throttling. Option A is incorrect because permanently increasing RCU to 2000 would waste capacity during the other 22 hours, increasing costs unnecessarily.

Option B is incorrect because DynamoDB Accelerator (DAX) adds cost and primarily caches base table reads, not global secondary index reads; the GSI is also throttled, so DAX may not fully resolve the issue, and scheduled scaling is more cost-effective for predictable patterns. Option D is incorrect because on-demand capacity mode is more expensive for predictable, sustained high traffic; auto scaling with a schedule is cheaper for this use case.

Page 21

Page 22 of 23

Page 23