A new database in the NoSQL category of AWS database suites, adding to a total of 12 database offerings from AWS.

What is a Memory DB?

As the name suggests Memory DB is a completely in-memory database (uses RAM, rather than an HDD/SSD) solution offered by AWS under the NoSQL category. In-memory databases are ideal for applications that require microsecond response times and can have large spikes in traffic coming at any time.

AWS memory DB was developed in rivalry with SAP HANA.

AWS Memory DB for Redis:

Redis is a compatible in-memory database that can completely replace the traditional database stead of using it as a traditional cache-store.

ElastiCache vs Memory DB

ElastiCache

- Acts as a plain key-value store based on TTL. Cannot be used as a primary database.
- Traditional cache systems are used to reduce no of reads to cache and minimize the no of calls that hits the database by caching them.
- Also used in places where 3rd-party (partner integrations) API integrations where the partner API's throttled, Caching used to save the API response in cache store for a period to save redundant requests to partner API

High level architecture:

Memory DB

- Can be used as a primary database that is completely working by storing the data in memory.
- Potentially replaces the primary database

High level architecture:

Use-cases of Memory-DB

- Caching.
- High volume transactional data.

So what are the other popular in-memory database? SAP HANA

Reference:

aws.amazon.com/nosql/in-memory aws.amazon.com/memorydb