Operational Excellence in the AWS Well-Architected Framework

leading up to @awscloud #reinvent, I've been diving into AWS resources & sharing what I find

first it was whitepapers, then AWS Labs code projects. this week, it’s the AWS Well-Architected Framework

everything (and more) up at https://markn.ca/2021/aws-reinvent/

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

there are 5 pillars in the framework, today, let's look at the Operational Excellence pillar of the AWS Well-Architected Framework

this thread is available unrolled at https://t.co/WcNfjEXkPV

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

this pillar is all about making sure that your architecture works and continues to work, regardless of what happens

yes, things happen, and yes, your system should continue to work when they do 😉

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

the AWS def. of this pillar is, "the ability to support development and run workloads effectively, gain insight into their operations, and to continuously improve supporting processes and procedures to deliver business value"

I prefer mine, less adjectives

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

this pillar really works in three phases:

1. organize and prepare
2. operate
3. evolve

ideally, this is a strong feedback loop & with each iteration, your team is getting better & your design is improving

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

six principles guide this pillar. the first is

All operations are code

• type it somewhere => script
• clicked on something => add it to a script/template
• make any change => find that line in code

be ruthless here…

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

...having every op as code (in addition to your infrastructure, security, and code) will save you a ton of headaches later

“code” is a catch-all here. it can be a script (bash + awscli), CDK, 3rd party tool, CFN template, Terraform, something repeatable

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

this leads to the next principles; update your documentation automatically

a lot of the “how” and “what” docs should be auto generated. that saves the people effort for the “why” docs

remember, writing this stuff is 1% of work, 99% is reading. docs are 🔑

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

3rd principle: small frequent changes

despite an initial concern, this actually reduces risk. you get used to making changes (and build tooling to support) but small changes are easier to troubleshoot, evaluate, and rollback if necessary

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

4th principle: iterate

…a lot

…like all the time

…feedback loops => small improvements => big changes over time!

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

💩 happens. no matter how much you plan & prepare. accept that. plan for that

understanding that recovery & adjustments are a key part of ops will lead to smarter decisions (#security too)

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

last principle: always learn. failure, sucess, complete disaster << they are all learning opportunities

the only failure is the failure to learn…or some other such cliche

just try not to make the same mistake twice. you can do that by learning 😉

🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

there's a lot more in the operational excellence pillar of the AWS Well-Architected Framework

you can read through the pillar at https://docs.aws.amazon.com/wellarchitected/latest/operational-excellence-pillar/welcome.html

/🧵☁️ #cloud #devops

@marknca tweeted at 22-Nov-2021, 17:33

👆 Start