Item: Amazon DynamoDB
Rating: 8
Author: Verified User

Use Cases and Deployment Scope

DynamoDB is used across the Engineering department. The biggest use case was introduced by me a couple of years back where our largest production workload was migrated to DynamoDB from Cassandra. More can be found here: https://techblog.gumgum.com/articles/moving-to-amazon-dynamodb-from-hosted-cassandra

Pros and Cons

Auto-scaling is the best feature which no other product offers.
Predictable performance with good latency metrics.
Its 100% server-less and fully managed which makes engineer's lives easier.

Global Table is not the best solution yet and needs to be improved.
Does not do well in the case of globally replicable counters.
Cost of GT might be excessive and not feasible for most companies with big production-write workloads.

Return on Investment

Cost savings over Cassandra.
Less of a maintenance hassle and less downtime for the services running on it.

Alternatives Considered

I authored the below blog which covers this in a much better way:

https://techblog.gumgum.com/articles/moving-to-amazon-dynamodb-from-hosted-cassandra

Other Software Used

Scylla, Cassandra, IntelliJ IDEA, MySQL, Amazon Elastic Compute Cloud (EC2), Amazon CloudWatch, Amazon ElastiCache

Likelihood to Renew

Predictable performance and it meets our SLA expectations.

Likelihood to Recommend

Amazon DynamoDB is well suited for high-traffic workloads which vary across the day. The autoscaling feature helps to reduce the running cost for off-peak hours.

It is not suitable for counters especially when they need to be replicable. The item size is also important. If it's trying to read too much, it might not give the best performance.

Amazon DynamoDB is an amazing, reliable NoSQL data store

Overall Satisfaction with Amazon DynamoDB