RDS is the easy, scalable option, provided you don't need to do too much voodoo
March 14, 2017

RDS is the easy, scalable option, provided you don't need to do too much voodoo

Michael E. Gruen | TrustRadius Reviewer
Score 8 out of 10
Vetted Review
Verified User

Overall Satisfaction with Amazon Relational Database Service

We use RDS (MySQL and Postgres) for important production systems in our backend data systems. It solves the problem of not having to maintain (patch, upgrade, clean) an installation of our databases, freeing up our DBA resources to work on more important things. Further, it is fast, easily scalable, and very reliable.
  • Easy to scale. Need more concurrent connections or more space? Reconfigure and relaunch. This allows us to start small and have room to scale.
  • Easy to maintain. While db exploits aren't as common as OS-level issues, they do come up. With RDS, upgrades happen automatically. It's quite nice.
  • Easy to work with. If you're not doing anything crazy with your database and just need it to work, RDS just works. Sometimes, your user permissions might need to be set a little oddly due to how RDS admins are set up, but it's a minor nit.
  • Not fully configurable. It's a hosted service and you sometimes needs to work around RDS's rules, especially when it comes to administrating the service. But, it's similar to AWS's general architecture, so not a huge issue.
  • Aurora is the "better" service from a performance perspective. From a cost perspective, RDS is fine and will likely improve over time.
  • Cheaper to run your own database on your own instance, but I'm really stretching for a con here as it's relatively so inexpensive.
  • Easy to get started: low time cost.
  • Scaling cost with usage — better dollar cost management
There really isn't a comparable service. Azure was surprisingly complicated to set up and crashed at odd points during a POC without much help. We looked at Rackspace to check it out, but as most of our infrastructure is in AWS, any benefit to Rackspace's offerings were nullified by keeping it all in one place.
If you can host your stuff in the cloud, AWS RDS is probably the best go-to that I've worked with to date. It is flexible enough for most applications, includes enough SQL dialects (MySQL, Postgres, et al) to satisfy any engineer who wants to work with a relational database for general purposes. It's also useful for smaller data warehousing. For larger applications and analytics warehousing, use Redshift or similar. RDS will fail/be slow. There are better, specialty products for this.