Learn AWS Application and Database

Operational invariants untuk sistem AWS application dan database: correctness, durability, idempotency, ordering, recovery, observability, alerting, dan runbook.

Failure-Model-First Design

Failure-model-first design untuk sistem AWS application dan database: taxonomy kegagalan, containment, graceful degradation, retry, circuit breaker, database failure, async replay, workflow recovery, dan production test matrix.

Reference Architecture Baseline

Reference architecture baseline untuk AWS application dan database: API boundary, command service, database transaction boundary, outbox, event bus, queue, workflow, projection, cache, observability, dan failure containment.

Integration Style Decision Tree

Decision tree untuk memilih integration style di AWS: synchronous API, queue, pub/sub, event bus, workflow, stream, read model, dan direct database boundary berdasarkan invariants, coupling, failure model, dan operability.

Synchronous Request Response on AWS

Synchronous request-response di AWS: API Gateway, AppSync, ALB, Lambda/ECS backend, latency budget, timeout, retry, throttling, backpressure, idempotency, transaction boundary, dan operability.

Async Messaging on AWS

Asynchronous messaging di AWS untuk membangun sistem yang tahan spike, partial failure, retry, duplicate delivery, poison message, fanout, dan backlog tanpa mengorbankan correctness.

Orchestration vs Choreography

Orchestration versus choreography di AWS: kapan memakai Step Functions, EventBridge, SNS/SQS, saga, state machine, domain events, compensation, dan operational control plane.

Command, Event, Query Separation

Command, event, dan query separation untuk sistem AWS: intent mutation, domain facts, read model, outbox, projection, consistency boundary, dan implementation blueprint.

Delivery Semantics: At-Most-Once, At-Least-Once, Effectively-Once

9 min

Delivery semantics di AWS: at-most-once, at-least-once, exactly-once claim, effectively-once application design, idempotency, deduplication, replay, DLQ, dan operational safety.

Idempotency, Deduplication, Replay, and Deterministic Side Effects

7 min

Idempotency, deduplication, dan replay-safe design di AWS untuk API, queue consumer, workflow, outbox/inbox, database transaction, dan external side effects.

Contract-First Integration: Schema, Versioning, Compatibility, Consumer Safety

6 min

Contract-first integration untuk API, event, queue message, workflow input, dan database projection di AWS dengan schema compatibility, versioning, validation, dan consumer safety.

API Gateway Deep Usage: REST API, HTTP API, WebSocket API

API Gateway deep usage untuk REST API, HTTP API, dan WebSocket API sebagai production boundary di atas application dan database workload AWS.

API Contract Design: OpenAPI, Request Validation, Error Shape, Compatibility

API contract design dengan OpenAPI, request validation, error model, compatibility rules, versioning, dan API Gateway integration extensions untuk production AWS systems.

Timeout, Retry, Rate Limit, Quota, and Throttling pada API Layer

9 min

Timeout, retry, rate limit, quota, throttling, and backpressure pada API layer AWS agar synchronous API tetap predictable di production.

Idempotent Command Handling di API Layer

Idempotent command handling di API layer AWS: Idempotency-Key, command store, request hash, transaction boundary, response replay, dan safe retry untuk mutation endpoints.

API-to-Database Transaction Boundary

Transaction boundary API-to-database di AWS: unit of work, isolation, optimistic concurrency, idempotency store, outbox, retry, partial failure, dan correctness invariant.

API Gateway Private Integration and VPC Link

API Gateway private integration dan VPC Link di AWS: public API boundary ke private service, ALB/NLB integration, routing, security, observability, failure modes, dan production checklist.

GraphQL AppSync in Action

AWS AppSync in action untuk GraphQL API, resolver, subscription, realtime, data source composition, dan database boundary yang aman di production.

API Observability and Debugging

API observability dan debugging untuk API Gateway, AppSync, Lambda/backend service, queue/event boundary, dan database dependency di AWS production system.

SQS Mental Model: Queue sebagai Load-Leveler, Buffer, dan Failure Isolation

SQS mental model sebagai queue, load-leveler, buffer, dan failure isolation layer untuk sistem application/database production.

Standard vs FIFO Queue: Ordering, Throughput, Deduplication, Message Group

Perbandingan mendalam SQS Standard dan FIFO Queue dari sisi ordering, throughput, deduplication, message group, concurrency, dan correctness design.

Visibility Timeout, DLQ, Redrive, Poison Message, dan Retry Envelope

Visibility timeout, dead-letter queue, redrive, poison message, retry envelope, retention, dan production playbook untuk Amazon SQS.

Consumer Design: Polling, Batch, Partial Batch Failure, Backpressure

Desain consumer SQS production-grade: polling, batch, partial batch failure, idempotency, backpressure, concurrency, database writes, dan shutdown safety.

SQS Database Worker Pattern

SQS database worker pattern untuk menulis ke database secara aman, idempotent, observable, dan tahan retry/replay di production.

SQS Fanout with SNS and EventBridge

Fanout SQS dengan SNS dan EventBridge untuk publish sekali, konsumsi banyak, isolasi failure per consumer, dan event routing yang tetap aman untuk database-backed systems.

SQS Large Message Pattern

Large message pattern untuk SQS dengan payload reference, S3 pointer, metadata consistency, lifecycle cleanup, idempotency, dan failure handling di production.

SQS Operability Playbook

Playbook operability SQS untuk monitoring lag, age, DLQ growth, redrive, replay, autoscaling signal, backpressure, dashboard, alarm, dan incident response.

SNS Pub/Sub Mental Model

Mental model Amazon SNS sebagai pub/sub fanout layer untuk komunikasi asynchronous application-to-application, termasuk topic design, subscription boundary, delivery semantics, SQS fanout, FIFO topic, DLQ, retry, observability, dan failure modes.

SNS Message Filtering

Desain Amazon SNS message filtering sebagai routing contract, termasuk filter policy scope, message attributes, message body filtering, AND/OR logic, numeric/string operators, schema compatibility, governance, testing, observability, dan anti-pattern.

EventBridge Event Bus Design

EventBridge event bus design untuk membangun event routing boundary yang scalable, auditable, dan aman berevolusi di sistem AWS production.

EventBridge Event Contract

EventBridge event contract design untuk source, detail-type, envelope, schema, versioning, compatibility, validation, consumer safety, dan replay-safe evolution.

EventBridge Rules, Pipes, Scheduler: Routing, Enrichment, Scheduled Command

Deep in-action guide to Amazon EventBridge Rules, Pipes, and Scheduler as routing, enrichment, and scheduled-command primitives for production application/database systems.

Archive and Replay: Reprocessing Events Without Breaking Invariants

Deep in-action guide to Amazon EventBridge Archive and Replay, focusing on replay-safe event-driven systems, recovery, backfill, validation, and invariant protection.

EventBridge Cross-Account dan Cross-Region Event Routing

Cross-account dan cross-region event routing dengan Amazon EventBridge untuk sistem multi-account, multi-region, regulated, dan production-grade.

Event-Driven Operability

Playbook operability untuk event-driven system dengan Amazon EventBridge, SNS, SQS, workflow, database projection, replay, DLQ, schema drift, dan debugging production.

Step Functions Mental Model: State Machine sebagai Durable Control Plane

Step Functions mental model sebagai durable control plane untuk workflow, saga, orchestration, callback, retry, compensation, dan database-safe long-running process di AWS.

Standard vs Express Workflow: Duration, Cost, Throughput, Auditability

Perbandingan Standard vs Express Workflows di AWS Step Functions dari sisi durability, duration, execution semantics, cost model, observability, throughput, auditability, dan pattern implementasi.

AWS States Language in Action

Amazon States Language in action: Task, Choice, Pass, Wait, Parallel, Map, Succeed, Fail, data transformation, service integration, and production workflow grammar.

Step Functions Error Handling

Step Functions error handling in action: Retry, Catch, timeout, heartbeat, terminal errors, compensation, saga failure paths, and production recovery design.

Service Integration Patterns

Step Functions service integration patterns in action: Request Response, Run a Job .sync, Wait for Callback with Task Token, IAM, timeout, idempotency, and production design.

Saga Orchestration

11 min

Saga orchestration in action with AWS Step Functions: reservation, compensation, timeout, semantic locking, idempotent participants, human escalation, and distributed consistency.

Step Functions Database Workflow

Step Functions database workflow in action: long-running business process, short local transactions, reservations, outbox, idempotency, callback token, and database-safe orchestration.

Workflow Observability and Cost

Workflow observability, execution history, Step Functions cost model, logging, tracing, redrive, and production debugging for AWS application/database workflows.

Database Selection Framework

Database selection framework for AWS application and database systems based on access patterns, consistency, latency, query shape, operational invariants, cost, and reversibility.

Workload Characterization

Workload characterization for AWS database and application systems, covering OLTP, cache, graph, document, time-series, search projections, concurrency, consistency, growth, and operability.

Data Modeling Invariants: Aggregate, Ownership, Cardinality, Mutation Frequency

Data modeling invariants untuk AWS application/database system: aggregate, ownership, cardinality, mutation frequency, identity, ordering, uniqueness, lifecycle, dan failure-proof data model.

Consistency Models: Strong, Eventual, Read-After-Write, Transactional Boundary

Consistency models di AWS: strong, eventual, read-after-write, transactional boundary, monotonic read, projection lag, global replication, dan practical design untuk Aurora/RDS/DynamoDB/cache/search/event-driven systems.

Relational vs Key-Value vs Document vs Graph vs Time-Series

25 min

Relational vs key-value vs document vs graph vs time-series database, dilihat dari access pattern, consistency, query shape, invariants, operability, dan failure modes di AWS.

Polyglot Persistence with Discipline: Bukan Microservice = Database Baru

Polyglot persistence yang disiplin di AWS: kapan memakai banyak database, kapan tidak, bagaimana menjaga source of truth, consistency, operability, governance, dan migration safety.

Data Ownership: Source of Truth, Projection, Read Model, Derived State

Data ownership dan source of truth di AWS: cara menentukan authoritative store, derived state, projection, read model, event ownership, auditability, reconciliation, dan failure handling agar sistem tidak berubah menjadi distributed data mess.

Database Architecture Decision Record: Template, Trade-Off, Reversibility

11 min

Database Architecture Decision Record untuk AWS: template, evidence, workload characterization, access pattern, consistency, migration, reversibility, cost, operability, security boundary, dan governance agar keputusan database bisa diaudit dan dievolusi.

RDS dan Aurora Mental Model: Managed Relational Database Bukan Sekadar VM DB

RDS dan Aurora mental model sebagai managed relational database system, bukan sekadar VM database. Membahas control plane, data plane, storage, failover, replicas, endpoint, connection, transaction boundary, dan operability.

Engine Selection: PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, Db2

Engine selection untuk Amazon RDS dan Aurora: PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, Db2, Aurora PostgreSQL, dan Aurora MySQL. Membahas decision drivers, workload fit, licensing, compatibility, features, operations, migration, dan failure modes.

Aurora Architecture: Storage, Replication, Writer, Reader, Failover

24 min

Aurora architecture in action: cluster volume, distributed storage, replication, writer/reader topology, endpoints, failover, read scaling, backup behavior, global database, I/O cost, and operational failure modes.

Schema Design for Aurora PostgreSQL: Index, Constraint, Partition, JSONB

Schema design for Aurora PostgreSQL in production: relational modeling, constraints, indexes, JSONB, partitioning, concurrency, zero-downtime change, auditability, and operational performance.

Transaction Isolation, Locking, Deadlock, MVCC, and Retry Strategy

Transaction isolation, locking, deadlock, MVCC, and retry strategy for Aurora PostgreSQL/RDS-backed application systems.

Connection Management, Pooling, RDS Proxy, Serverless Clients, and Exhaustion

Connection management, application-side pooling, RDS Proxy, serverless connection storms, pinning, failover behavior, and production observability.

Read Scaling, Replicas, Reader Endpoint, Lag, and Read-Your-Writes Problem

Read scaling with Aurora/RDS replicas, reader endpoint semantics, replica lag, read-your-writes problem, endpoint strategy, routing patterns, and production observability.

Backup, Restore, PITR, Snapshot, Clone, and Disaster Recovery Drill

Backup, restore, point-in-time recovery, snapshots, Aurora clone, backtrack, recovery drills, RPO/RTO thinking, and production-grade database recovery strategy.

Aurora Serverless v2

Aurora Serverless v2 in action: scaling semantics, ACU capacity, pause/resume, pooling, failover, workload fit, cost traps, and production readiness.

RDS Operability and Performance Insights

RDS and Aurora production operability: Database Insights, Performance Insights migration, DB load, wait events, slow queries, alarms, runbooks, and incident analysis.

Aurora DSQL Mental Model

Aurora DSQL mental model: serverless distributed SQL, PostgreSQL compatibility boundary, strong consistency, optimistic concurrency control, automatic sharding, and application design implications.

Active-Active SQL Architecture

Active-active SQL architecture with Aurora DSQL: multi-Region consistency, endpoint routing, regional failure handling, latency trade-offs, conflict avoidance, and production design patterns.

DSQL Data Modeling

Aurora DSQL data modeling for transactional distributed SQL: transaction boundaries, primary key distribution, optimistic concurrency, schema compatibility, query discipline, and production-safe patterns.

DSQL Application Patterns

Application patterns for Aurora DSQL: multi-Region writes, endpoint routing, idempotent commands, OCC retry, conflict avoidance, outbox, workflow integration, cache/projection strategy, and failover behavior.

DSQL Operability: Lifecycle, Connectivity, Observability, Incident Drill

Aurora DSQL operability for production systems: monitoring, quotas, OCC conflicts, active-active runbooks, schema change, incident response, and readiness checks.

DynamoDB Mental Model: Partitioned Key-Value Store with Predictable Access

DynamoDB mental model for production systems: partitioned key-value/document storage, predictable access patterns, consistency, capacity, hot keys, item collections, and application correctness.

Single-Table Design: Entity, Access Pattern, Composite Key, Item Collection

Single-table design DynamoDB untuk production system: access pattern, entity modeling, composite key, item collection, sparse index, denormalization, versioning, dan operational safety.

Partition Key Design: Hot Key, Cardinality, Write Distribution, Adaptive Capacity

Partition key design DynamoDB untuk production: cardinality, hot key, write distribution, adaptive capacity, sharding, multi-tenancy, time-series, GSI hot partitions, dan failure modelling.

GSI and LSI Strategy: Query Shape, Projection, Sparse Index, Overloading

Strategi GSI dan LSI DynamoDB untuk production: query shape, projection, sparse index, index overloading, consistency, capacity, hot index, backfill, dan operability.

Transactions and Condition Writes: Optimistic Concurrency, Uniqueness, Counters

11 min

DynamoDB transactions dan condition writes untuk production correctness: optimistic concurrency, uniqueness, idempotency, counters, state transitions, conflict handling, dan retry strategy.

DynamoDB Streams: Outbox, Projection, Event Sourcing Adjacent Pattern

DynamoDB Streams untuk change data capture, projection, outbox-adjacent pattern, audit trail, event-driven integration, dan replay-safe stream processing.

Global Tables: Multi-Region Replication, Conflict Semantics, Locality

DynamoDB Global Tables untuk multi-Region, multi-active replication: MREC, MRSC, conflict semantics, locality, routing, write ownership, failure handling, dan production readiness.

DynamoDB Capacity and Cost

DynamoDB capacity and cost engineering: on-demand, provisioned, autoscaling, warm throughput, maximum throughput, TTL, storage growth, index amplification, global tables, and production cost controls.

DynamoDB Production Debugging

Production debugging playbook for DynamoDB: throttling diagnosis, hot partitions, conditional write failures, GSI backpressure, stream lag, backfill safety, retries, dashboards, and incident runbooks.

Cache Mental Model

Cache mental model for AWS application and database systems: latency optimization, derived state, consistency liability, cache-aside, write-through, TTL, jitter, stampede control, invalidation, failure modes, and production readiness.

ElastiCache Redis Valkey Patterns

ElastiCache for Valkey and Redis OSS in action: serverless vs node-based, cache-aside, write-through, TTL, jitter, invalidation, data structures, cluster topology, connection management, memory, eviction, failover, security, observability, and production patterns.

MemoryDB: Durable In-Memory Store, When Cache Becomes Primary State

MemoryDB sebagai durable in-memory database: kapan cache berubah menjadi primary state, bagaimana desain data, consistency, failover, idempotency, dan operability-nya di production.

Cache Invalidation: Versioned Key, Event-Driven Eviction, Stale-While-Revalidate

Cache invalidation sebagai distributed correctness problem: TTL, versioned key, event-driven eviction, stale-while-revalidate, negative caching, ordering, replay, dan failure mode di AWS.

Session State, Rate Limiting, Distributed Locking, and Token Bucket

Session state, rate limiting, distributed locking, token bucket, lease, fencing token, dan low-latency coordination state dengan ElastiCache, MemoryDB, DynamoDB, API Gateway, dan application code.

Cache Operability: Eviction, Memory Pressure, Stampede, Hot Key, Failover

Cache operability untuk ElastiCache/Redis/Valkey: eviction, memory pressure, stampede, hot key, failover, connection storm, observability, runbook, dan production readiness.

Amazon DocumentDB: Document Modeling, Compatibility Boundary, Query Pattern

Amazon DocumentDB in action: document modeling, compatibility boundary, query and index design, transactions, change streams, schema evolution, migration, operability, dan production readiness.

Amazon Neptune: Graph Modeling, Traversal, Relationship-Heavy Workloads

Amazon Neptune in action: graph modeling, property graph vs RDF, Gremlin/openCypher/SPARQL, traversal design, relationship-heavy workloads, query tuning, loading, operability, dan production readiness.

Amazon Keyspaces: Cassandra-Compatible Modeling, Partition, Clustering, Query

Amazon Keyspaces in action: Cassandra-compatible modeling, partition and clustering key design, query-first schemas, capacity, consistency, multi-Region behavior, operability, and production failure modes.

Amazon Timestream: Time-Series Modeling, Retention, Dimension, Measure

Amazon Timestream in action: time-series modeling, dimensions, measures, retention, ingestion, late data, scheduled queries, rollups, query patterns, operational dashboards, and production failure modes.

OpenSearch as Query Projection

OpenSearch as a query projection for search, filtering, analytics-adjacent queries, and operational dashboards without turning it into the source of truth.

Specialized Database Selection Cases

Specialized database selection case studies for graph, time-series, search, document, wide-column, cache, and composite regulatory workloads.

Database Migration Strategy

Database migration strategy untuk AWS: rehost, replatform, refactor, CDC, dual write, phased cutover, rollback, reconciliation, dan failure-mode-first migration planning.

AWS DMS in Action

AWS DMS in action: source/target endpoints, replication instance/serverless, full load, CDC, task settings, table mappings, validation, monitoring, cutover, and rollback.

Zero-Downtime Schema Change

Zero-downtime schema change dengan expand-migrate-contract, backward/forward compatibility, online backfill, dual-read/write discipline, validation, rollback, and production runbook.

Production Readiness Final Review