The Axiom

Cloud

59 pages

Start here

Hub

Cloud Engineering Brain

Central hub for all cloud and platform engineering knowledge. Every page in the Cloud brain connects here.

Advanced Load Balancing on AWS

ALB, NLB, Global Accelerator, and traffic management patterns for high-availability architectures.

albnlbglobal-acceleratortarget-groups

Amazon Bedrock AgentCore Runtime

AWS's serverless hosting platform for AI agents. GA October 2025. Each session runs in a dedicated microVM with isolated CPU/memory. Pay-per-second billing — no charge during LLM/tool I/O wait. Supports LangGraph, Strands, CrewAI, and any Python framework.

awsbedrockagentcoreserverless

Amazon Bedrock Guardrails

Amazon Bedrock Guardrails — managed content safety layer for FM applications. Six policy types: content filters, denied topics, word filters, sensitive information redaction, grounding checks (hallucination), and contextual grounding. AIF-C01 Domain 5 core.

awsbedrockguardrailscontent-filtering

Amazon EKS — Elastic Kubernetes Service

Managed Kubernetes on AWS: control plane management, node groups, Fargate profiles, and IRSA.

ekskubernetesnode-groupsfargate-profiles

Amazon EventBridge

Serverless event bus for decoupling producers from consumers across AWS services, SaaS, and custom applications.

eventbridgeevent-busrulespipes

Amazon Q (Business and Developer)

Amazon Q is AWS's GenAI assistant family — Q Business (enterprise GenAI over company data, permission-aware) and Q Developer (AI coding assistant). AIF-C01 tests Q Business vs Kendra: Q Business returns synthesised answers; Kendra returns document passages.

awsamazon-qamazon-q-businessamazon-q-developer

Amazon SageMaker Sub-Services

SageMaker sub-services — Canvas (no-code ML), Autopilot (AutoML), Ground Truth (data labelling), Data Wrangler (feature prep), Feature Store, Pipelines (MLOps), Model Monitor (drift), Clarify (bias/explainability), JumpStart (foundation models). AIF-C01 Domain 3 core.

awssagemakersagemaker-studiocanvas

Ansible

Agentless configuration management and automation tool. Uses SSH to push configuration to remote hosts — no daemon, no agent installed on targets. Written in Python; tasks are YAML playbooks.

ansibleiacconfiguration-managementplaybooks

Argo Rollouts

Progressive delivery controller for Kubernetes. Extends Deployments with canary, blue-green, and analysis-driven rollout strategies.

argo-rolloutsprogressive-deliverycanaryblue-green

ArgoCD

GitOps continuous delivery for Kubernetes. Watches a Git repository; reconciles cluster state to match. The cluster tells Argo what's running; Git tells Argo what should be running.

argocdgitopskubernetescd

AWS Analytics Services

AWS analytics services decision guide — Athena (serverless SQL on S3), EMR (managed Spark), Glue (serverless ETL), Kinesis (real-time streaming), OpenSearch (log search), QuickSight (BI), Redshift (data warehouse).

awsanalyticsathenaemr

AWS API Gateway

Fully managed API layer. Routes HTTP requests to Lambda, ECS, EC2, or any HTTP backend.

awsapi-gatewayrest-apihttp-api

AWS CDK

AWS Cloud Development Kit. Define AWS infrastructure in TypeScript, Python, Java, or Go.

aws-cdkcdkiaccloudformation

AWS Core Services

The foundational services every cloud engineer needs to know. AWS dominates cloud (33% market share, 2026). These ten services underpin the vast majority of production architectures.

awscloudec2s3

AWS ECS — Elastic Container Service

AWS-native container orchestration. Simpler than Kubernetes — no control plane to manage, no YAML manifests.

awsecsfargatecontainers

AWS ElastiCache

Managed Redis and Memcached in AWS. ElastiCache handles node provisioning, patching, failure detection, and replication. Use Redis for most workloads — richer data structures, persistence, pub/sub.

elasticacheredismemcachedaws

AWS Fargate

Serverless compute engine for containers — run ECS or EKS workloads without managing EC2 instances.

fargateecsserverless-containerstask-definition

AWS Lambda — Patterns and Production

Lambda beyond hello-world. Cold starts, concurrency, event sources, and deployment patterns that matter for production workloads.

awslambdaserverlessevent-driven

AWS Lambda Powertools

Production-grade observability, idempotency, and batch processing for Lambda — the standard library for serious Lambda work.

lambdapowertoolsloggertracer

AWS Networking — Advanced

Beyond basic VPC: multi-VPC architectures, transit gateway, PrivateLink, WAF, DNS failover, and network performance patterns.

awsnetworkingvpctransit-gateway

AWS Pre-Built AI Recognition Services

AWS pre-built AI services — Rekognition (vision), Comprehend (NLP), Lex (chatbots), Polly/Transcribe (speech), Translate (language), Textract (documents), Kendra (search), Personalize (recommendations), Forecast (time series). No ML expertise required.

awsairekognitioncomprehend

AWS RDS & Aurora

Managed relational database services. RDS manages MySQL, PostgreSQL, MariaDB, Oracle, SQL Server.

awsrdsaurorapostgresql

AWS SageMaker

Managed ML platform: training, model registry, real-time inference endpoints, and batch transform — the bridge between cloud engineering and AI engineering.

sagemakerml-servingendpointstraining-jobs

AWS Security and Compliance

Security Hub, GuardDuty, Config, Inspector, and WAF — the AWS security control plane.

security-hubconfigguarddutyinspector

AWS Service Ecosystem

Overview of the AWS service ecosystem for AI practitioners — compute, storage, AI/ML, observability, and security services.

awscloudecosystemsagemaker

AWS SQS and SNS

The messaging backbone of AWS event-driven architectures. SQS = queue (point-to-point). SNS = pub/sub (one-to-many fan-out).

awssqssnsmessaging

AWS Step Functions

Serverless orchestration for distributed workflows. Coordinates Lambda functions, ECS tasks, SQS, SNS, DynamoDB, and 200+ AWS services into reliable state machines.

awsstep-functionsstate-machineorchestration

Azure Core Services

Microsoft Azure. Third cloud globally but dominant in enterprise (Microsoft 365 integration, Entra ID as de facto corporate identity).

azuremicrosoftaksazure-functions

Blue-Green and Advanced Deployment Strategies

Zero-downtime deployment patterns with instant rollback capability.

blue-greendeploymentcanaryrolling

CDN and Edge Patterns

Content delivery networks and edge computing: moving content and computation closer to users to reduce latency and origin load. Critical for global applications.

cdncloudfrontedgecaching

Cloud Migration

Planning and executing workload migrations to AWS: the 6 Rs, wave planning, and the tools that make it work.

migration6rslift-and-shiftre-architect

Cloud Monitoring

Observability for cloud infrastructure. You cannot improve what you cannot measure. The three pillars: metrics (what is happening), logs (why it happened), traces (where time was spent).

monitoringobservabilitycloudwatchprometheus

Cloud Networking

Network design underpins security, performance, and cost in cloud architectures. The same concepts — VPC, subnets, routing, firewalls — appear in all three major clouds with different names.

networkingvpcsubnetsload-balancing

Cloud Security

Securing AWS infrastructure: IAM least privilege, preventive controls (SCPs, resource policies), detective controls (GuardDuty, CloudTrail, Security Hub), and network security (WAF, Security Groups, N...

awssecurityiamguardduty

Cloud-Native Patterns

Design principles and patterns for applications built to run on cloud infrastructure: containerised, dynamically orchestrated, microservices-oriented, and built for scale and resilience.

cloud-nativetwelve-factorsidecarservice-mesh

Container Security

Securing the container lifecycle: from image build to runtime in Kubernetes. Containers reduce attack surface compared to VMs but introduce their own threat model.

container-securitydockerkubernetestrivy

Data Engineering on Cloud

Building data pipelines and analytics infrastructure on AWS. From S3 data lake to queryable analytics with Athena, Glue, and Redshift.

data-engineerings3glueathena

Disaster Recovery

Planning and executing recovery from catastrophic failures — region outages, data corruption, ransomware, accidental mass deletion.

disaster-recoveryrtorpobackup

Docker

Container runtime. Packages an application and its dependencies into a portable, reproducible image that runs identically in dev, CI, and production.

dockercontainersdockerfilemulti-stage

FinOps and Cloud Cost Management

Engineering discipline for understanding, controlling, and optimising cloud spend.

finopscosttaggingbudgets

GCP Core Services

Google Cloud Platform. Second cloud for most teams; first cloud for ML/AI workloads and analytics. GCP leads on Kubernetes (invented it), BigQuery, and Vertex AI. 12% market share (2026).

gcpgoogle-cloudgkecloud-run

GitHub Actions

CI/CD built into GitHub. Triggered by events (push, PR, schedule, manual). Runs jobs in parallel or sequence across GitHub-hosted or self-hosted runners.

github-actionsci-cdoidccaching

GitOps Patterns

GitOps uses Git as the single source of truth for infrastructure and application state. A GitOps operator continuously reconciles the actual cluster state toward what's declared in Git.

gitopsargocdfluxprogressive-delivery

Helm — Advanced Patterns

Beyond `helm install`. Hooks, library charts, OCI registries, testing, and patterns for production-grade chart management.

helmkuberneteschartshooks

Infrastructure Monitoring on AWS

CloudWatch, X-Ray, Synthetics, and SLO-driven alerting for AWS workloads.

cloudwatchmonitoringdashboardssynthetics

KEDA — Kubernetes Event-Driven Autoscaling

KEDA (Kubernetes Event-Driven Autoscaling) extends Kubernetes HPA to scale workloads based on external event sources — Kafka consumer lag, SQS queue depth, Prometheus metrics, Redis list length, and 6...

kedakubernetesautoscalingkafka

Kubernetes

Open-source container orchestration. Declare desired state; the control plane makes it so and keeps it so. The standard production substrate for containerised workloads.

kubernetesk8spodsdeployments

Kubernetes Operators

Operators extend Kubernetes with domain-specific knowledge.

kubernetesoperatorscrdskubebuilder

Multi-Tenancy Patterns

Serving multiple customers from shared infrastructure.

multi-tenancysaastenant-isolationdata-partitioning

Observability Stack

The three pillars of observability: metrics (what's broken), logs (why it broke), traces (where it broke).

observabilityprometheusgrafanaloki

Platform Engineering

Building and operating an Internal Developer Platform (IDP) that enables product teams to self-serve infrastructure, deployments, and tooling — without needing deep ops expertise.

platform-engineeringinternal-developer-platformbackstagegolden-paths

Pulumi

Infrastructure as Code using real programming languages. Write AWS/GCP/Azure resources in TypeScript, Python, Go, or .NET.

pulumiiactypescriptpython

Quality Gates in GitOps Delivery

Quality gates in a GitOps flow split across two distinct planes — CI gates (PR checks before a commit lands in Git) and GitOps gates (sync hooks, analysis runs, and promotion policies that execute after Git already holds the desired state). QA practitioners own both planes.

gitopsquality-gatesci-cdargocd

Secrets Management

Credentials, API keys, database passwords, TLS certificates — anything that grants access must be managed, rotated, and audited.

secretsvaulthashicorpaws-secrets-manager

Serverless Patterns

Architectural patterns for serverless compute — Lambda, Cloud Run, Azure Functions.

serverlesslambdacloud-runevent-driven

Service Mesh

A service mesh handles cross-cutting network concerns for microservices: mTLS, retries, circuit breaking, traffic shaping, and observability — without touching application code.

service-meshistiolinkerdenvoy

Terraform

Infrastructure as Code tool by HashiCorp (now part of IBM, with community fork OpenTofu under MPL 2.0). Declare infrastructure in HCL, run plan-apply-destroy.

terraformiacinfrastructure-as-codehcl

VPC Design Patterns

Network architecture patterns for secure, scalable AWS deployments — from single-account to multi-account.

vpcsubnetsnat-gatewaytransit-gateway