Not a fan of RDS. It has all the same drawbacks of using an EBS partition for storage, but as far as I know you have no visibility (like await, iowat etc) to know when your storage subsystem starts to flake out. EBS removes a lot of Sysadmin & DBA headaches, but when you suddenly get mysterious DB stalls, good luck trying to figure out what's wrong.
I've worked with some systems that used a software RAID10 array of EBS volumes, and when a single EBS volume started to flake out, they'd just kick it out of the array, and performance would improve. They'd then rebuild the failed replica during off hours. If you're going to rely on EBS for DB storage for anything in production, this seems like this is the way to go.
I second this. While EC2 takes a bunch of really hard problems, allowing you to provision virtual servers at a moment's notice, the value proposition for RDS is much more marginal. It's simply not that hard to get MySQL up and running, even in a RAIDed configuration, and the fact that the underlying store is EBS means most of the advantage of RAID is spent simply overcoming the I/O handicaps of EBS.
We prefer running our MySQL instances on ephemeral storage, which is physically closer to the instance, and deal with I/O volumes by using lots of replication.
The article about the new MacBook Pro line also included a discussion of Airs and Mac Pros. Sometimes the discussion strays beyond the boundaries of the headline.
Most of the time that is a good thing - I for one am interested in the point the OP raised, which is that RDS (at all instance sizes) gives you no visiblity in to what is happening under the covers on your DB server.
He's informing us of the drawbacks of using the other plans. This special plan would be (for some) dipping toes in the water of RDS only to find out its infested with piranhas.
Yeah, I have to wonder if there's an exclusivity agreement with the Oracle licensing in RDS that requires that MySQL and Oracle be the only options for linux databases.
I will try one for a few weeks and see what kind of performance it provides.
I was hoping that Heroku's new PostgreSQL service would be a good option for a hosted easy to use database but I have lots of small projects and Heroku's offerings are per database not per virtualized database server.
I could potentially use a hosted postgres solution, i just tried to go to heroku's site to look at what they've got. their homepage is currently an error message saying "no application is configured for this hostname". That certainly inspires confidence.
This is a great update. You certainly don't want to put anything production on here, but small applications would be great with this. Even a properly cached WP blog (low traffic) would do just fine on here.
The automatic backups and rollback in RDS is worth way more than $19 alone.
I sure wish they would allow the slaving of RDS instances to non-RDS instances (aka CHANGE MASTER TO). I would love to host a MySQL slave at Amazon AWS.
Sure, I can get this done with an EC2 instance but how killer would a "DR slave" with rollbacks and automatic backup features be? Besides feature-sets I could eliminate quite a few delayed replication nodes and dedicated DR servers at hosted providers if they would offer this feature.
This is actually a small example of how Amazon keeps being aggressively competitive by making the pricepoint almost impossible for anyone else to match at scale.
That's an odd way to restate what the original poster said. Your argument appears to be the tautology that AWS is expensive when compared to different services which do less.
First, that's not a tautology. Second, that's not my point. The OP is wrong in saying that Amazon are being price-competitive. That's not their strategy; you can see it by comparing their VM costs to just about anyone else out there. Amazon's strategy is to out-feature everyone and lock customers in precisely so that they don't have to compete on price.
It is the definition of a tautology: you're both saying the same thing but from different directions. AWS is trying to avoid room for equivalent services, not services which do less: they don't want a customer who's sensitive to $2/month but they do want people to say “ops is expensive, let's save our people for non-commodity jobs”
Also, those cost comparisons almost universally compare spot pricing rather than reserved. If you run the numbers seriously AWS is not the bargain basement leader but it's a lot more even than the contrarian blog echo chamber would have you believe, particularly if you have bursty workloads, use some of their other offerings and/or need servers outside of the US/Europe.
Currently for one of my sites we have the web server and MySQL database on one medium instance. I assume performance would be a lot worse on a micro rds and same web box?
I've worked with some systems that used a software RAID10 array of EBS volumes, and when a single EBS volume started to flake out, they'd just kick it out of the array, and performance would improve. They'd then rebuild the failed replica during off hours. If you're going to rely on EBS for DB storage for anything in production, this seems like this is the way to go.