1. A website's page load times are gradually increasing as more users access the system at the same time. Analysis indicates that a user profile is being loaded from a database in all the web pages being visited by each user and this is increasing the database load and the page load latency. To address this issue the Developer decides to cache the user profile data. Which caching strategy will address this situation MOST efficiently?

A) Use a dedicated Amazon RDS instance for caching profile data. Use a write-through caching strategy.
B) Create an Amazon ElastiCache cluster to cache the user profile data. Use a cache-aside caching strategy.
C) Create an ElastiCache cluster to cache the user profile data. Use a write-through caching strategy.
D) Create a new Amazon EC2 Instance and run a NoSQL database on it. Cache the profile data within this database using the write-through caching strategy.



2. A company maintains an application responsible for processing several thousand external callbacks each day. The company's System administrators want to know how many callbacks are being received on a rolling basis, and they want this data available for 10 days. The company also wants the ability to issue automated alerts if the number of callbacks exceeds the defined thresholds. What is the MOST cost-effective way to address the need to track and alert on these statistics?

A) Push callback data to an Amazon RDS database that can be queried to show historical data and to alert on exceeded thresholds.
B) Push callback data to Amazon Kinesis Data Streams and invoke an AWS Lambda function that stores data in Amazon DynamoDB and sends the required alerts.
C) Push callback data to AWS X-Ray and use AWS Lambda to query, display, and alert on exceeded thresholds.
D) Push callback data to Amazon CloudWatch as a custom metric and use the CloudWatch alerting mechanisms to alert System Administrators.



3. Queries to an Amazon DynamoDB table are consuming a large amount of read capacity. The table has a significant number of large attributes. The application does not need all of the attribute data. How can DynamoDB costs be minimized while maximizing application performance?

A) Batch all the writes, and perform the write operations when no or few reads are being performed.
B) Create a global secondary index with a minimum set of projected attributes.
C) Implement exponential backoffs in the application.
D) Load balance the reads to the table using an Application Load Balancer.



4. A company has multiple Developers located across the globe who are updating code incrementally for a development project. When Developers upload code concurrently, internet connectivity is slow and it is taking a long time to upload code for deployment in AWS Elastic Beanstalk. Which step will result in minimized upload and deployment time with the LEAST amount of administrative effort?

A) Create an AWS CodeCommit repository, allow the Developers to commit code to it, and then directly deploy the code to Elastic Beanstalk.
B) Allow the Developers to upload the code to a central FTP server to deploy the application to Elastic Beanstalk.
C) Allow the Developers to upload the code to an Amazon S3 bucket, and deploy it directly to Elastic Beanstalk.
D) Create a code repository on an Amazon EC2 instance so that all Developers can update the code, and deploy the application from the instance to Elastic Beanstalk.



5. An application stops working with the following error: The specified bucket does not exist. Where is the BEST place to start the root cause analysis?

A) Check AWS CloudTrail for a DeleteBucket event.
B) Check the application logs in Amazon CloudWatch Logs for Amazon S3 DeleteBucket errors.
C) Check the Elastic Load Balancer logs for DeleteBucket requests.
D) Check AWS X-Ray for Amazon S3 DeleteBucket alarms.