The question I'm trying to answer now: If the business were to invest in a classic, Oracle data warehouse, what do we lose STRATEGICALLY by NOT investing in modern cloud solutions?
Or what's the trade-off?
What's the opportunity-cost?
We'll come back to that.
Just some quick notes, off the top of my head:
RDS = Relational Database Service
It's a suite of cloud database services offered by Amazon. Including:
- SQL Server - you know it
- Oracle - I dislike SQL Developer...MS SQL Developer is so much better
- MySQL - quick and easy. I've used it for quick POC projects.
- Aurora - fast, serverless, and hyper scalable. The way of the future. Maybe Amazon's answer to BigQuery?
- Postgres - Open source and supposedly become preferred over SQL Server lately. Since some Microsoft strategy backfired.
- Redshift - Massive SQL Data Warehouse.
These all use SQL.
There are probably more.
Amazon also offers non-traditional/NoSQL database services, such as:
- Mongodb - Uses a JavaScript/JSON-like querying.
- Neptune - Graph database service.
- Redis - An in-memory cache. For blazing-fast performance.
There are more...I'll get to that.
One of the neat things Amazon tries to do, is to separate storage (actual hard drives somewhere out in the world) from the actual database engines.
So, this means you can collect a lot of data in a lot of different ways, but still put it all in one place. So that these many different database services can then be layered on top.
This gives us the opportunity to use the best tool for the job.
And THIS is the what I've been missing.
My team is currently using Google BigQuery to power some of our POC solutions. But we used it because it's what we already had. We don't have the budget and cross-functional alignment to change that yet.
But to do this right, we need to make sure we're using the right tool for the job.
Need to build a recommendation engine or a social network? Use Neptune. It's a graph database. Want something for user profiles, content management, catalogs? You need a document database. Check out Amazon DocumentDB.
Sticking to a single solution will hold us back.
The right tool for the job.
Amazon has a nice chart.
I'm going to take Amazon's course on purpose build databases.
It's already clear to me that cloud database solutions are superior.
As a Product Manager, the question I'm trying to answer now: If the business were to invest in a classic, Oracle data warehouse, what do we lose STRATEGICALLY by NOT investing in modern cloud solutions?
Or rather: what is the net value (beyond just dollars)
I should try for a comparison chart, looking at these features:
- Maximum # of Concurrent Users/DB Connections/Data Consumers
- Maximum Database Storage Size
- Cost of Storage
- Cost of Compute
- Agility
More on this as I learn...