Amazon Web Services Updated DVA-C02 Exam Questions and Answers by aura

A Lambda alias is a pointer to a specific Lambda function version or another alias1. A Lambda alias allows you to invoke different versions of a function using the same name1. You can also split traffic between two aliases by assigning weights to them1.

In this scenario, the developer needs to test their changes in production on a small percentage of all traffic without changing the API Gateway URL. To achieve this, the developer can follow these steps:

Define an alias on the $LATEST version of the Lambda function. This will create a new alias that points to the latest code of the function.

Update the API Gateway endpoint to reference the new Lambda function alias. This will make the API Gateway invoke the alias instead of a specific version of the function.

Upload and publish the optimized Lambda function code. This will update the $LATEST version of the function with the new code.

On the production API Gateway stage, define a canary release and set the percentage of traffic to direct to the canary release. This will enable API Gateway to perform a canary deployment on a new API2. A canary deployment is a software development strategy in which a new version of an API is deployed for testing purposes, and the base version remains deployed as a production release for normal operations on the same stage2. The canary release receives a small percentage of API traffic and the production release takes up the rest2.

Update the API Gateway endpoint to use the $LATEST version of the Lambda function. This will make the canary release invoke the latest code of the function, which contains the optimized changes.

Publish to the canary stage. This will deploy the changes to a subset of users for testing.

By using this solution, the developer can test their changes in production on a small percentage of all traffic without changing the API Gateway URL. The developer can also monitor and compare metrics between the canary and production releases, and promote or disable the canary as needed2.

DynamoDB distributes throughput across partitions based on the hash key. A hot partition (caused by high usage of a specific hash key) can result in a ProvisionedThroughputExceededException, even if overall usage is below the provisioned capacity.

Why Option B is Correct:

Partition-Level Limits: Each partition has a limit of 3,000 read capacity units or 1,000 write capacity units per second.

Hot Partition: Excessive use of a single hash key can overwhelm its partition.

Why Not Other Options:

Option A: DynamoDB storage size does not affect throughput.

Option C: Provisioned scaling operations are unrelated to throughput errors.

Option D: Sort keys do not impact partition-level throughput.

DynamoDB Partition Key Design Best Practices

The described behavior is a classic cache stampede (also called the “thundering herd” problem): a large set of hot keys share the same expiration time, so they all expire simultaneously. When that happens, many requests miss the cache at the same moment and fall back to the backend database, creating a sudden spike in database load and causing increased latency and poor user experience.

The most effective fix that preserves freshness is to introduce jitter —a small, randomized variance—into the TTL values so cached objects do not all expire at the same time. With TTL jitter, each item still has a bounded lifetime (so freshness is maintained), but expirations become spread out over time. That smooths cache-miss rates and prevents synchronized regeneration, reducing bursts of database traffic. This is a common best practice in high-load caching systems precisely to avoid mass expiration events.

Option A (increase TTL) might reduce how frequently a stampede occurs, but it does not solve the root cause: keys can still align and expire together, and a longer TTL can also reduce freshness if product data changes. Option B (more replicas, disable cluster mode) changes topology and read scaling, but it does not prevent synchronized expiry or the resulting backend spike; disabling cluster mode could also reduce scale for large keyspaces. Option D (more shards) can improve cache throughput capacity, but it does not address the core issue that backend load spikes occur when items expire simultaneously; lowering replica count could even reduce read availability.

Therefore, C is the correct solution: keep TTL-based freshness, but add a randomized component (for example, TTL = baseTTL + random(0..jitter)) so expirations are distributed and the backend database is protected from sudden bursts.