QuestionQ53

Resilient Cloud Solutions

A company runs an application that tracks user activity on the company’s website and mobile apps. The application uses Amazon ElastiCache for Redis as a write-through cache and an Amazon RDS for PostgreSQL database for longer-term storage. When the application receives a request to record a user action, it writes simultaneously to the Redis cluster and the database. Internal recommendation applications consume this data to produce content recommendations for each user.

During peak periods, the recommendation applications cannot produce user recommendations because data is stale or missing. The Redis cache has cluster mode disabled, and the database has a single read replica.

The company wants to ensure that the recommendation applications can generate content recommendations during peak periods. A DevOps engineer has already created a new ElastiCache for Redis cluster with cluster mode enabled.

What should the DevOps engineer do next to meet the company’s requirements?

Explanation

ElastiCache for Redis cluster-mode-enabled clusters support sharding, and a target tracking policy using ElastiCachePrimaryEngineCPUUtilization dynamically adds or removes shards to maintain the selected utilization target as load changes. The configuration endpoint lets a cluster-aware client discover the primary and read endpoints of every shard, allowing the recommendation applications to access the distributed Redis data rather than a single replica endpoint.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!