Exam AWS Certified Developer Associate topic 1 question 280 discussion

It is for sure D. Write-through is better for data that is changing often. But here it is not the case. The app is doing READS a lot, and few WRITES. That means LAZY LOADING is better. When it comes to TTL, postcode and geolocation is not something that changes frequently. So, you can go for larger TTL. It saves cost, cache is not invalidated too often, and you will not need to update cache very often. Also, your requests will be very fast, since it will directly read from cache. Shortly, the answer is D.

Cache evicts data with small TTL value. So choosing large TTL value.

In my opinion C is the correct anwer. First of all because using write-through caching strategy the writes are so heavy but guarantee that the cache is always updated (it is ok because there are more reads than writes). Secondly, it is ok to choose a lazy loading strategy (D) but I don't understand why you guys put a TTL record in the table. We don't want to remove the row from the table.

For those arguing that there are more reads than writes, this argument just makes option C the most optimised solution as it improves latency by doing less DB roundtrips as the cache is not stale. Option D will make the data stale in the cache and will eventually increase latency once the cache is invalidated.

C is correct answer. Write-through caching strategy - is appropriate for an application that performs many more reads than writes, and the application retrieves the same data repeatedly. By using a write-through caching strategy, the application writes data to the cache and the database simultaneously. This ensures that the data in the cache is always up-to-date and the database is used only to retrieve data that is not in the cache. This approach can improve the application's read performance and decrease latency. Write-behind caching strategy - is appropriate for an application that performs more writes than reads. It is a technique that defers database updates until after new data is written to the cache. In the given scenario, as the application performs many more reads than writes, this strategy is not suitable.

Go with C

I chose C. The reason is that with write-through caching strategy, data will always be available at cache, so there will be less latency. One of the disadvantages of using write-through is related to situations when you cache data that will never be read (Cache churn), but this questions states that "The application retrieves each postal code from the database thousands of times each day", so most of the data cached will be read.

Why not C? It says each postcode will be accessed thousands of times a day. Also, the content won't change very often, because the coordinates of a zipcode is unlikely to change.