AWS Certified Developer - Associate Flashcards

Card 1 of 300 mastered
Say the answer out loud before flipping.
Browse all 30 cards
  1. What is the difference between a Lambda function's concurrency and a reserved concurrency limit?

    Concurrency is the number of in-flight executions a function has at any moment. Reserved concurrency sets both a guaranteed minimum and a hard maximum for a specific function, preventing it from consuming the account's shared pool or from scaling beyond that cap.

  2. When should you use an SQS FIFO queue instead of a standard queue?

    Use FIFO when message order must be preserved and duplicates cannot be tolerated, such as processing financial transactions in sequence. Standard queues offer higher throughput but only best-effort ordering and at-least-once delivery.

  3. What does an SQS dead-letter queue (DLQ) do?

    It captures messages that fail processing after a configured number of receive attempts (maxReceiveCount), isolating problem messages so they don't block the main queue and allowing them to be inspected or reprocessed later.

  4. What is the purpose of a Lambda execution role?

    It is the IAM role that grants a Lambda function permission to interact with other AWS services (e.g., writing to S3 or DynamoDB); it is separate from any role used to invoke the function.

  5. How does DynamoDB eventually consistent read differ from strongly consistent read?

    Eventually consistent reads may return stale data shortly after a write but consume half the read capacity and have lower latency; strongly consistent reads always return the most recent write but cost more read capacity and are unavailable for global secondary indexes.

  6. What is a DynamoDB Global Secondary Index (GSI) used for?

    A GSI lets you query a table using an alternate partition key (and optional sort key) different from the table's primary key, enabling additional query patterns without duplicating the table.

  7. What triggers a DynamoDB Stream record?

    An item-level modification (insert, update, or delete) in a DynamoDB table, which can then be consumed by Lambda or the Kinesis Client Library to react to data changes in near real time.

  8. What is the AWS Encryption SDK / envelope encryption pattern used by KMS?

    Data is encrypted with a data encryption key (DEK), and the DEK itself is encrypted with a KMS customer master/managed key (CMK); this avoids sending large payloads to KMS and lets KMS manage only key wrapping.

  9. What is the difference between an IAM role and an IAM user?

    An IAM user represents a persistent identity with long-term credentials; an IAM role is an identity with temporary credentials that can be assumed by users, services, or applications, and is the recommended way to grant permissions to compute resources.

  10. What does AWS STS AssumeRole return?

    Temporary security credentials (access key, secret key, and session token) that expire after a set duration, allowing cross-account or cross-service access without long-lived credentials.

  11. What is the purpose of an S3 pre-signed URL?

    It grants time-limited access to a specific S3 object using the permissions of the URL's creator, allowing a client to upload or download the object directly without needing AWS credentials of their own.

  12. What S3 storage class is best for infrequently accessed data that must remain immediately retrievable?

    S3 Standard-Infrequent Access (S3 Standard-IA), which offers lower storage cost than S3 Standard with millisecond retrieval, but charges a retrieval fee and has a minimum storage duration.

  13. What does S3 multipart upload accomplish?

    It splits a large object into parts uploaded independently (and in parallel), improving throughput and allowing resumption of a failed upload without restarting the entire transfer.

  14. What is CodeDeploy's AppSpec file used for?

    It defines deployment configuration for EC2/on-premises or Lambda deployments, including source file locations, lifecycle event hooks, and (for Lambda) the function version and alias to shift traffic to.

  15. What is the difference between CodeDeploy's in-place and blue/green deployment types?

    In-place deployment stops the application on existing instances and installs the new version on them; blue/green provisions a new set of instances (or resources) with the new version and shifts traffic over, allowing quick rollback by reverting traffic.

  16. What is a Lambda alias, and why use it with weighted traffic shifting?

    An alias is a named pointer to a specific Lambda function version; assigning weighted routing across two aliases enables canary or linear deployments that gradually shift invocation traffic to a new version.

  17. What problem does API Gateway usage plans and API keys solve?

    They let you throttle and meter API consumption per client, enforcing rate limits, burst limits, and quotas for individual API keys tied to specific customers or applications.

  18. What is the purpose of AWS X-Ray in a distributed application?

    It traces requests as they travel through multiple services, producing a service map and timing data that helps developers identify latency bottlenecks and pinpoint errors across microservices.

  19. How do X-Ray annotations differ from metadata?

    Annotations are indexed key-value pairs that can be used to filter and search traces; metadata is not indexed and is used only to record additional contextual detail on a trace segment.

  20. What is the difference between CloudWatch Logs and CloudWatch Metrics?

    CloudWatch Logs stores raw log event data emitted by applications and services; CloudWatch Metrics stores numeric time-series data (like CPU utilization) that can trigger alarms and be graphed on dashboards.

  21. What does a CloudWatch metric filter do?

    It scans incoming log data for a specified pattern and converts matches into a custom CloudWatch metric, enabling alarms to be built on values that only appear inside log text.

  22. What is the purpose of AWS Systems Manager Parameter Store versus Secrets Manager?

    Parameter Store holds configuration values and can optionally encrypt them with KMS, at little or no cost; Secrets Manager is purpose-built for secrets like database credentials, offering automatic rotation and tighter secret-lifecycle features at a higher cost.

  23. What is idempotency in the context of API design, and how can DynamoDB conditional writes support it?

    Idempotency means repeating the same request produces the same result without unintended side effects. A DynamoDB conditional write (e.g., attribute_not_exists) can reject duplicate writes for the same idempotency key, preventing duplicate processing of a retried request.

  24. What is the difference between synchronous and asynchronous Lambda invocation?

    Synchronous invocation waits for the function to complete and returns the response directly to the caller (e.g., API Gateway); asynchronous invocation queues the event and returns immediately, with Lambda handling retries and optional destinations for success/failure.

  25. What is a Lambda destination, and how does it differ from a DLQ?

    A destination is a target (SQS, SNS, Lambda, or EventBridge) that receives a record about an asynchronous invocation's outcome, and can be configured separately for success and failure; a DLQ only captures failed events and provides less detail than a destination.

  26. What causes a Lambda cold start, and how can provisioned concurrency help?

    A cold start occurs when Lambda must initialize a new execution environment (download code, start the runtime, run init code) before handling a request, adding latency. Provisioned concurrency pre-initializes a specified number of environments so they are ready to respond immediately.

  27. How does Amazon SNS message filtering work?

    Subscribers attach a filter policy to their subscription; SNS evaluates the policy against message attributes and only delivers messages that match, allowing a single topic to fan out selectively to different subscribers.

  28. What is the difference between CloudFormation change sets and stack updates?

    A change set previews the proposed additions, modifications, and deletions a template update would make without executing them; a stack update applies those changes directly, which can be riskier since there's no preview step.

  29. What does CORS (Cross-Origin Resource Sharing) enforcement require on an API Gateway resource?

    The API must respond to preflight OPTIONS requests with appropriate Access-Control-Allow-* headers, and the actual method responses must also include the allowed origin header, or browsers will block the client-side request.

  30. What is the difference between the AWS SDK's default retry behavior and exponential backoff?

    AWS SDKs automatically retry throttled or transient failures using exponential backoff with jitter, progressively increasing the wait time between retries to avoid overwhelming a service that is already struggling under load.