Reliability
8 min read
How we test failover every Thursday
Every Thursday at 10am Eastern we fail over one Postgres cluster in each region on purpose. Here is what broke the first six times.
Written by
Sam Okafor
Every Thursday at 10am Eastern we fail over one Postgres cluster in each region on purpose. Here is what broke the first six times.
This is sample article text for the template. Replace it with your own post in the CMS: every paragraph, heading and list here is a normal rich text field.
What we measured
We pulled a year of invoices, split every line into compute, storage and traffic, and compared each account against its first full month on Aldridge.
Compute cost fell by a median of 22%
Storage cost was roughly flat
Traffic cost went to zero because egress is included
What we would do again
Keep the old stack running for a week, move one service at a time, and switch DNS last. Most of the risk in a migration is in doing everything at once.
Keep reading
Pricing
What our egress bill would have been
We ran the numbers on 1,140 accounts that moved from AWS. The median one was paying 31% of its bill for traffic leaving the cloud.
Company
Why we built in Columbus
Cheap power, a fiber crossroads and a lot of people who know how to run buildings. Our first region was not an accident.
Pricing
Per-second billing, explained with one GPU
A walk through one worker-gpu-1 that ran for 5 hours, 52 minutes and 11 seconds, and exactly what it cost.