Enabling hybrid operations on AWS

Slide 1

Slide 1 text

Enabling hybrid operations on AWS Darko Meszaros Developer Advocate - AWS @darkosubotica ln/darko-mesaros twitch.tv/ruptwelve youtu.be/ruptwelve

Slide 2

Slide 2 text

What will we talk about today? • What is Hybrid? • What are Operations? • Operations in a Hybrid Cloud • Best Practices

Slide 3

Slide 3 text

Slide 4

Slide 4 text

Agility Complexity and risk Benefits of cloud Accelerate time to market Innovation Scale seamlessly Optimize costs Minimize security vulnerabilities Reduce management complexity

Slide 5

Slide 5 text

Why do customers need hybrid cloud? Run workloads on-premises Run workloads in the cloud Provide tight integration Leverage skills and resources

Slide 6

Slide 6 text

Hybrid cloud stats $96B Source: IDC, Rightscale, Forrester, Markets & Markets 71% 75% On-premises Cloud Hybrid cloud Size of hybrid market by 2023 Organizations with hybrid deployments Enterprise applications that are still on-premises

Slide 7

Slide 7 text

Hybrid architecture goals Operational consistency Existing skillsets and tools Simple to control, manage, secure Enterprise-class app SLA Build once, deploy anywhere

Slide 8

Slide 8 text

Data center extension Cloud bursting Backup and disaster recovery to the cloud Application migration Geographic expansion

Slide 9

Slide 9 text

$(whoami) Darko Mesaroš / Darko Meszaros / Дарко Месарош @darkosubotica ln/darko-mesaros twitch.tv/ruptwelve youtu.be/ruptwelve

Slide 10

Slide 10 text

Slide 11

Slide 11 text

So, what do you think are operations?

Slide 12

Slide 12 text

Operations with agility Optimize Analyze and reduce cost, improve efficiency and security posture Act Take operational action on resources Audit Audit resource configurations, user access, and policy enforcement Monitor Monitor resources and applications

Slide 13

Slide 13 text

Operations with agility Monitor resources and applications Optimize to reduce cost and improve security posture Manage resources and take operational action Audit user activity and resource configurations Amazon CloudWatch AWS Trusted Advisor AWS Cost and Usage Report AWS Cost Explorer AWS Systems Manager AWS CloudTrail AWS Config

Slide 14

Slide 14 text

Observability is key to success

Slide 15

Slide 15 text

Amazon CloudWatch Complete observability of cloud resources and applications Monitor Visualize applications and infrastructure, correlate logs and metrics, set alerts Act Automate responses to operational changes Collect Metrics and logs form resources, applications, and services on AWS and on premises Analyze Real-time analysis with CloudWatch Metric Math and CloudWatch Logs Insights

Slide 16

Slide 16 text

A central point from which to manage operations

Slide 17

Slide 17 text

Any environment Operate any AWS or external resource centrally Open Agent is open-sourced on GitHub Multi-platform Windows and Linux support Automated Multi-account, multi-Region automation AWS Systems Manager Centrally manage hybrid cloud resources at any scale

Slide 18

Slide 18 text

Enough talk – I want action!

Slide 19

Slide 19 text

Slide 20

Slide 20 text

Perform operations as code • Apply the same engineering discipline that you use for application code to your entire environment. • Define your entire workload (applications, infrastructure, etc.) as code. • Make updates through code

Slide 21

Slide 21 text

Make frequent, small, reversible changes • Design workloads to allow components to be updated regularly • Make changes in small increments that can be reversed. • These changes – if they fail – should not affect your customers.

Slide 22

Slide 22 text

Refine operations procedures frequently • Look for opportunities to improve operational procedures as you do them • Evolve your workload -> evolve your operations. • Playbooks vs Runbooks • Game Days are fun!

Slide 23

Slide 23 text

Anticipate failure • Pre-Mortem excercises • Test your failure scenarios. • Test your response procedures! • Game Days are still fun!

Slide 24

Slide 24 text

Pagers and Notifications • Have an action per notification! • Do not page/notify someone unless there is something to do. • Notify the right person

Slide 25

Slide 25 text

Things to note: • Well-Architected framework • DevOps Hanbook - book • https://aws.amazon.com/hybrid/