RousselTMLEARNING

Cloud computing

AWS best practices: security, reliability, costs and operations

The essential best practices for running AWS in production: multi-account organization, identities and least privilege, networking, encryption, high availability, backups, observability, infrastructure as code and cost control.

1. Organization and identities #

The foundations.

Rules #

AWS landing zone
Each account has a role; SCPs set common guardrails.
  • Several accounts (at least production, non-production, security, logs) under Organizations.
  • Root user protected by MFA and never used; human access through Identity Center.
  • Roles for applications, no access keys in the code.
  • SCPs: allowed regions, forbidding disabling CloudTrail or GuardDuty.

2. Security and reliability #

Protecting and holding up.

Rules #

  • Resources in private subnets; only load balancers and CloudFront are exposed.
  • Encryption everywhere (S3, EBS, RDS) with KMS, TLS in transit.
  • Secrets in Secrets Manager, never in the code or in plain-text variables.
  • Everything across at least two AZs; tested AWS Backup backups; RTO and RPO defined.
  • GuardDuty, Security Hub, Config and CloudTrail active everywhere.

3. Operations and costs #

Lasting without drifting.

Rules #

  • All infrastructure as code, deployed through a pipeline; no manual changes in production.
  • Mandatory tags (project, environment, owner) for cost allocation.
  • Budgets and anomaly detection on every account.
  • CloudWatch alarms on user symptoms (errors, latency), log retention defined.
  • Annual Well-Architected review of critical workloads.
aws sts get-caller-identity
aws iam get-account-summary --query 'SummaryMap.AccountMFAEnabled'
aws guardduty list-detectors
aws securityhub get-findings --filters '{"SeverityLabel":[{"Value":"CRITICAL","Comparison":"EQUALS"}]}' --max-items 5
aws s3control get-public-access-block --account-id 123456789012
aws ce get-cost-forecast --time-period Start=2026-10-28,End=2026-11-30 --metric UNBLENDED_COST --granularity MONTHLY

Related articles

Observability on AWS: CloudWatch, X-Ray, Application Signals, ADOT, Managed Prometheus and Grafana

A complete overview of observability on AWS: CloudWatch components (metrics, logs, alarms, Application Signal…

Understanding AWS: global infrastructure, accounts, tools and pricing

The essential basics for getting started with Amazon Web Services: regions, availability zones and points of …

IAM: identities, roles and policies on AWS

Secure access to AWS with IAM: users, groups, roles and JSON policies, permission evaluation, MFA, IAM Identi…

Keep learning

Support us

♥ SUPPORT US

Support RousselTM Learning

Courses, articles, labs and certificates stay free with no intrusive ads. A voluntary contribution, of any amount you choose, pays for hosting, lab environments and new content.

Contribute with PayPal Any amount · card or PayPal accountLearn more
QR code PayPal
Or scan this QR code with your phone

Thank you, every contribution counts!