Amazon Redshift

1. It is a part of the AWS ecosystem. So in order to involve other AWS products, you don't have to go through APIs, oAuth credentials, or anything of the sort. The integration is much more streamlined. For example, depending on how your databases are setup, you can reference data outputs from S3 or Lambda logs easily. 2. The cost is much lower when compared to competitors and the tooling Redshift provides allows you to predict cost much easier than its competitors. 3. Since it's an analytical database, using Redshift as the source for dashboards allows for faster loading of data and running of queries. Since we do the majority of our transformations in dbt, a lot of our queries are synthesized into tables which means that we use Redshift for semantic layer transformations which is what most users are frustrated with when they refer to long loading times for dashboards. Using Redshift makes this not as much of an issue.

ease of connection

Scalability as it has been able to scale with our busy period when we see increased traffic. Easy of use, the query language is a modified SQL which means our analysts can use it without too much bother. Documentation, the platform has plenty of robust documentation and a active user community.

1. Creating and updating incremental tables can be very difficult due to Redshift's table lockout mechanism. 2. Navigating the UI (in terms of schemas, columns, CRUD functions) can feel outdated and frustrating as it is not intuitive. 3. Permissions and access can be a bit unwieldy as it's not intuitive.

rbac across late binding views adding sort keys to large existing tables recreating large tables to update dist keys

The only real issue I have is that the SQL dialect it uses is too close to others, meaning I trip up frequently when writing queries and I usually have to write them twice.