Exam AWS Certified Solutions Architect - Associate SAA-C02 topic 1 question 401 discussion

C. Deploy a two-node Amazon ElastiCache cluster and modify the application to query the cluster first and query the database only if needed.

- Disk I/O is NOT the bottleneck here --> "no significant performance improvement" after upgrading to higher-IOPS storage - Query planning and execution (of long-running, expensive queries) is the problem A. Migrate the RDS instance to an Amazon Redshift cluster and enable weekly garbage collection. - Wrong. Redshift is a data warehouse solution for OLAP (analytical) workloads. We are dealing with OLTP (transactional) workloads B. Separate the long-running queries into a new Multi-AZ RDS database and modify the application to query whichever database is needed. - Wrong. How are you going to synchronize these 2 different databases? C. Deploy a two-node Amazon ElastiCache cluster and modify the application to query the cluster first and query the database only if needed. - Correct. Cache results of long-running queries, and hit the cache for identical future requests D. Create an Amazon Simple Queue Service (Amazon SQS) FIFO queue for common queries and query it first and query the database only if needed. - Wrong. SQS isn't used as a DB substitute

"common queries run for long durations and memory utilization is high" this leads us directly to ElastiCache

Issue here is the memory, so deploying an AWS Elastic Cache cluster for the application to query it first before query the DB if need be, that way the DB performance will increase.

Memory is the problem. So caching is the best solution

Agreed with C. it stated "memory is significant". need fix cache

C. Deploy a two-node Amazon ElastiCache cluster and modify the application to query the cluster first and query the database only if needed.

A . Did anyone notice the mention of '700 IPOS' & 'long running queries'. This means query volume is normal but queries are complex and memory intensive. A typical criteria for redshift. Also note that it's a shopping cart application, where you write often than you read. So cache many not be as helpful. So C may not be as suitable.

A . Did anyone notice the mention of '700 IPOS' & 'long running queries'. This means query volume is normal but queries are complex and memory intensive. A typical criteria for redshift.

Answer is C-Elasticache. Elasticache helps reduce load off of databases for read intensive workloads NOTE-Using ElastiCache involves heavy application code changes A- Redshift is more for big data analytics/datawarehousing B-how will the application know which one to query C-fits well D-SQS is not used for querying,it is used to decouple applications

Anytime you see anything along the lines of "common queries/reads" immediately think "caching" as a solution.

Two node elastic cache cluster

This is about the performance issue. ElastiCache definately helps, hence C.

it's C 100%

Answer is A. https://docs.aws.amazon.com/redshift/latest/mgmt/working-with-clusters.html#working-with-clusters-overview An Amazon Redshift data warehouse is a collection of computing resources called nodes, which are organized into a group called a cluster. Each cluster runs an Amazon Redshift engine and contains one or more databases.

"common queries run for long durations" C is the answer; ElastiCache

C to me.