3 ms·
For those that have done this, I'd be interested in the performance gain (say, IOPS at an equivalent hosting + hardware cost). I'd also be interested in RDS vs
by samcheng 10y ago
For those that have done this, I'd be interested in the performance gain (say, IOPS at an equivalent hosting + hardware cost). I'd also be interested in RDS vs. Heroku Postgres.
Do people do this for security / policy purposes, or are they motivated by cost savings? Does removing all that virtualization buy improved performance?
- vacri 10y agoI'm not sure that the article moved to bare iron. The author could have just set up postgres on a regular ec2 instance.
- samcheng 10y agoDoes that bring performance or cost savings? What is the motivation and calculus for making that transition?
- eliribble 10y agoIt can bring a cost savings, but the savings is very slight if you're comparing apples-to-apples. For example, if you're using RDS in a multi-AZ replication set up with periodic S3 backups you have to take into account running each of the replica instances and the S3 costs. What we found was that RDS was treated like a black box to our automation (which was a good thing for a long time) and so we were less attentive to being over-provisioned and actually testing failover situations than we should have been.
- eliribble 10y agoYes, we did move to a regular EC2 instance for our cloud service. Where we install databases on premises for our customers it can be bare iron, but it can also be a similar virtualized setup.
- qaq 10y ago30,000 iops is less then single crappy consumer grade SSD 6TB is fairly low limit largest instance has 244Gb RAM So improved performance, running larger instances, lower cost