I’d like to present you the book I am trying to finish reading for some time now; a very dense book, with good practices and interesting details on how to keep planet-wide systems up & running with a bunch of very well prepared people.
What are the lessons one needs to walk away with, from this book? A few bullets:
I have mentioned before that regular EBS storage in AWS works with a system of credits that are accrued when the volume is idle and are spent when the operation rate requested on the volume grows over the predefined baseline. One can find the detailed explanation in the AWS documentation. This limitation came as an unexpected surprise during the failed project I was part of 18 months back.
A few days ago while syncing data from a large volume to another, larger volume, I ran into the same problem as well and I decided to capture the state to make things clear for everybody (well, people that are reading this blog).
The CloudWatch read throughput (metric name: VolumeReadOps) graph looked like this during the data sync:
AWS provides a complete monitoring engine called CloudWatch; it works with metrics – including custom, user-provided metrics – and is able to raise alarms when any such metric crosses a certain threshold. This is the tool that is used for all perfomance monitoring tasks within AWS.
This text will cover a monitoring scenario regarding deploying an arbitrary appplication to the cloud and being able to determine what causes the performance limits to be met, be it the application code itself or resource limits enforced by Amazon.
Scenario
Let’s assume that one has just started using Amazon Web Services and is deploying applications on free tier or other general purpose (T2) instances. One learns that the general purpose instances work with “credits” that allow dealing with short spikes through performance bursting – but once the credits are exhausted the performance is reverted to some baseline. All the particular details do not make a lot of sense but one needs to know if the application can meet the desired service limits with this setup.