Skip to content
Distributed Systems & Microservices Architecture
Search
Ctrl
K
Cancel
GitHub
Select language
English
Türkçe
I. Foundations & The Laws of Physics
Part Overview
1.1. From Monoliths to Distributed Systems
Chapter Overview
1.1.1. The Scaling Dilemma: Vertical vs. Horizontal Scaling
1.1.2. Cost-Performance Curves and Hardware Limits
1.1.3. Stateful vs. Stateless Services: When to Use Which
1.1.4. The 8 Fallacies of Distributed Computing
1.1.5. Network Reliability, Bandwidth Limitations, and Topology Changes
1.2. System Models and Failure Types
Chapter Overview
1.2.1. Network Models: Synchronous, Asynchronous, and Partially Synchronous
1.2.2. Failure Models: Crash-Stop, Crash-Recovery, Omission
1.2.3. Byzantine Faults: Malicious or Corrupted Actors
1.2.4. Deterministic vs. Probabilistic Failure Models
1.2.5. Failure Propagation: Cascading Failure Analysis
1.3. Time and Ordering
Chapter Overview
1.3.1. Physical Clocks: Quartz Drift, NTP Limitations, Leap Second Problem
1.3.2. Monotonic Clocks vs. Time-of-Day: Which One to Use
1.3.3. The Happened-Before Relation: Foundation of Causality
1.3.4. Lamport Timestamps: Capturing Causality
1.3.5. Vector Clocks: Detecting Concurrency
1.3.6. Hybrid Logical Clocks (HLC): Combining Physical and Logical Time
1.3.7. Google TrueTime & Spanner: Working with Uncertainty
1.3.8. Distributed ID Generation: UUIDv7, Snowflake, ULID
II. Communication Protocols & API Design
Part Overview
2.1. Synchronous Communication Protocols
Chapter Overview
2.1.1. REST: Constraints, Resources, and HTTP Contracts
2.1.2. HTTP/2 and HTTP/3: Multiplexing, HOL Blocking, and QUIC
2.1.3. gRPC & Protocol Buffers: High-Performance RPC
2.1.4. gRPC Streaming: Unary, Server-Streaming, Bi-directional
2.1.5. GraphQL: Query Language and Federation Architecture
2.1.6. WebSocket and Server-Sent Events: Real-Time Communication
2.1.7. API Versioning Strategies
2.2. Asynchronous Messaging Fundamentals
Chapter Overview
2.2.1. Message Queue vs. Event Streaming: Differences and Use Cases
2.2.2. Kafka Architecture: Topic, Partition, Consumer Group, Offset
2.2.3. Pulsar vs. Kafka: Architectural Comparison
2.2.4. RabbitMQ: Exchange Types and Routing
2.2.5. At-Least-Once, At-Most-Once, and Exactly-Once Semantics
2.2.6. AsyncAPI: Documenting Asynchronous Contracts
2.2.7. Schema Evolution and Schema Registry: Avro, Protobuf, and Compatibility Modes
2.2.8. Dead Letter Queues and Poison Message Handling
2.3. API Gateway and Service Contracts
Chapter Overview
2.3.1. API Gateway Pattern: Single Entry Point, Routing, Auth
2.3.2. Backend for Frontend (BFF) Pattern
2.3.3. OpenAPI / Swagger: Contract-First Development
2.3.4. Consumer-Driven Contract Testing (Pact)
2.3.5. Idempotency and Safe HTTP Methods
2.3.6. Throttling and Rate Limiting at the Gateway Level
2.4. Load Balancing
Chapter Overview
2.4.1. L4 vs. L7 Load Balancing: Transport vs. Application Layer
2.4.2. Algorithms: Round Robin, Least Connections, Weighted, Consistent Hashing
2.4.3. Client-Side vs. Server-Side Load Balancing
2.4.4. Global Server Load Balancing (GSLB) and Anycast
2.4.5. Health-Aware Routing and Connection Draining
III. Data Management & Consistency
Part Overview
3.1. Data Partitioning / Sharding
Chapter Overview
3.1.1. Key-Range Partitioning: Advantages and Range Scans
3.1.2. Hash Partitioning: Uniform Distribution
3.1.3. Skew and Hot-Spot Problems: Real-World Scenarios
3.1.4. Consistent Hashing and Virtual Nodes (vnodes)
3.1.5. Dynamic Rebalancing: Zero-Downtime Strategies
3.1.6. Colocated Joins: Placing Related Data on the Same Shard
3.1.7. Secondary Indexes in Partitioned Data: Local vs. Global
3.2. Replication Strategies
Chapter Overview
3.2.1. Leader-Based Replication: Synchronous vs. Asynchronous
3.2.2. Failover and Leader Election: The Split-Brain Risk
3.2.3. Replication Lag Problems: Read-Your-Own-Writes
3.2.4. Multi-Leader Replication: Write Conflicts and Resolution
3.2.5. Leaderless Replication (Dynamo-style): Quorum R + W > N
3.2.6. Anti-Entropy: Read Repair and Merkle Trees
3.3. Consistency Models
Chapter Overview
3.3.1. Linearizability: The Strongest Guarantee, the Highest Cost
3.3.2. Serializability: Transaction Isolation
3.3.3. Weak Isolation and MVCC: Read Committed, Snapshot Isolation, Write Skew
3.3.4. CAP Theorem: CP vs. AP Systems
3.3.5. PACELC Theorem: The Latency-Consistency Trade-off
3.3.6. Causal Consistency and Session Guarantees: The Middle of the Spectrum
3.3.7. Eventual Consistency: Last Write Wins (LWW) and Its Weaknesses
3.3.8. CRDTs: Conflict-free Replicated Data Types
3.4. Storage Engines and Data Structures
Chapter Overview
3.4.1. LSM-Tree vs. B-Tree: Write and Read Performance Trade-offs
3.4.2. SSTables, Memtable, and Compaction Strategies
3.4.3. Bloom Filters: Reducing Disk Reads
3.4.4. Write-Ahead Log (WAL): The Foundation of Durability
3.4.5. Columnar Storage: Advantages in Analytical Queries
3.5. Distributed Caching
Chapter Overview
3.5.1. Cache Strategies: Cache-Aside, Read-Through, Write-Through, Write-Behind
3.5.2. Cache Invalidation: TTL, Event-Based, and Versioned Keys
3.5.3. Redis Cluster Architecture: Sharding, Replication, Sentinel
3.5.4. Cache Stampede and Thundering Herd: Prevention Techniques
3.5.5. Distributed vs. Local Cache: Consistency Trade-offs
3.5.6. Memcached vs. Redis: Architectural Comparison
IV. Data Processing Architectures
Part Overview
4.1. Batch Processing
Chapter Overview
4.1.1. The MapReduce Paradigm: Functional Roots, Map, Shuffle, Reduce
4.1.2. Distributed Sort: The Challenge of the Shuffle Phase
4.1.3. From Hadoop to Apache Spark: Disk vs. In-Memory Processing
4.1.4. RDD, DataFrame, and Dataset APIs
4.1.5. Checkpointing and Fault Tolerance
4.1.6. Straggler Tasks: Speculative Execution
4.2. Stream Processing
Chapter Overview
4.2.1. Bounded vs. Unbounded Data
4.2.2. Event Time vs. Processing Time: The Critical Distinction
4.2.3. Windowing: Tumbling, Sliding, and Session Windows
4.2.4. Watermarks: The Art of Managing Late Data
4.2.5. Triggers and Accumulation Modes
4.2.6. Apache Flink: Depth of Stateful Stream Processing
4.3. Hybrid Architectures and the Log Abstraction
Chapter Overview
4.3.1. The Commit Log Abstraction: The Core Idea Behind Kafka/Pulsar
4.3.2. Lambda Architecture: The Complexity of Batch + Speed Layers
4.3.3. Kappa Architecture: Everything Is a Stream
4.3.4. Log Compaction: Kafka Compacted Topics and Tombstones
4.3.5. Change Data Capture (CDC): Real-Time Synchronization
4.3.6. Debezium: Practical Database CDC Implementation
V. Coordination & Consensus
Part Overview
5.1. Distributed Transactions
Chapter Overview
5.1.1. ACID vs. BASE: Guarantees and Trade-offs
5.1.2. Two-Phase Commit (2PC): Blocking Nature and Risks
5.1.3. Three-Phase Commit (3PC): Reducing Blockage
5.1.4. The SAGA Pattern: Choreography vs. Orchestration
5.1.5. Compensating Transactions
5.1.6. The Outbox Pattern: Atomic DB Write + Message Publishing
5.2. Consensus Algorithms
Chapter Overview
5.2.1. The Split-Brain Problem and Fencing Tokens
5.2.2. The FLP Impossibility Result: Why Consensus Is Hard
5.2.3. Paxos: Core Logic, Prepare and Accept Phases
5.2.4. Raft: Leader Election, Log Replication, Membership Changes
5.2.5. Zab (ZooKeeper Atomic Broadcast): Epochs and Zxid
5.2.6. etcd: Distributed K-V Storage on top of Raft
5.2.7. Byzantine Fault Tolerance: PBFT, Proof of Work / Proof of Stake
5.3. Distributed Locking
Chapter Overview
5.3.1. Lock Fundamentals: Mutual Exclusion in Distributed Systems
5.3.2. Lease-Based Locking: TTL, Renewal, and Expiry Semantics
5.3.3. Fencing Tokens: Preventing Stale Lock Holders
5.3.4. Redlock Algorithm: The Redis Distributed Lock Debate
5.3.5. ZooKeeper and etcd Recipes: Production Lock Implementations
VI. Microservice Architecture Patterns
Part Overview
6.1. Service Decomposition Strategies
Chapter Overview
6.1.1. Domain-Driven Design (DDD): Bounded Context and Ubiquitous Language
6.1.2. Drawing Service Boundaries: Conway's Law
6.1.3. The Strangler Fig Pattern: Safe Migration from Monolith
6.1.4. Decompose by Business Capability vs. Subdomain
6.1.5. Anti-Pattern: The Distributed Monolith - How to Avoid It
6.1.6. Service Granularity: How Small Is Too Small?
6.1.7. The Modular Monolith: When Not to Distribute
6.2. Structural Patterns
Chapter Overview
6.2.1. The Sidecar Pattern
6.2.2. The Ambassador Pattern: Helper Container as Proxy
6.2.3. The Adapter Pattern: Standardized Interfaces
6.2.4. Anti-Corruption Layer (ACL): Integration with Legacy Systems
6.2.5. Shared Library vs. Sidecar: Dependency Management Trade-offs
6.3. Service Discovery
Chapter Overview
6.3.1. Client-Side Discovery: Eureka, Ribbon, and Spring Cloud LoadBalancer
6.3.2. Server-Side Discovery: Load Balancer + Registry
6.3.3. Self-Registration vs. Third-Party Registration
6.3.4. Consul: Service Discovery + Health Checking + KV Store
6.3.5. DNS-Based Discovery: Kubernetes DNS
6.3.6. Health Check Endpoint Pattern
6.4. Data Management Patterns
Chapter Overview
6.4.1. Database per Service: Independence and Isolation
6.4.2. Shared Database Anti-Pattern: Why It Is Dangerous
6.4.3. API Composition: The Distributed Join Problem
6.4.4. Polyglot Persistence: The Right Database for the Right Job
6.4.5. Cross-Service Query: The Materialized View Pattern
6.5. Event Sourcing and CQRS
Chapter Overview
6.5.1. Event Sourcing: The Event Chain Instead of State
6.5.2. Event Store Design: The Append-Only Log
6.5.3. CQRS: Separating Read and Write Models
6.5.4. Projections: Building Read Models from Events
6.5.5. Event Versioning and Schema Evolution
6.5.6. Snapshots: Optimizing Long Event Chains
VII. Resilience & Production Engineering
Part Overview
7.1. Flow Control and Overload Management
Chapter Overview
7.1.1. Queueing Theory Foundations: Little's Law and the Utilization Curve
7.1.2. Backpressure: Pushing Load Back to the Source
7.1.3. Load Shedding: Which Requests to Drop and How
7.1.4. Priority Queues: Protecting Critical Requests
7.1.5. Rate Limiting: Token Bucket, Leaky Bucket, Fixed and Sliding Window
7.1.6. Adaptive Rate Limiting: Dynamic Limits
7.1.7. Retry Storms and Metastable Failures: Work Amplification and Retry Budgets
7.2. Protection Patterns
Chapter Overview
7.2.1. Circuit Breaker: The Closed to Open to Half-Open Cycle
7.2.2. Bulkhead Pattern: Thread Pool and Semaphore Isolation
7.2.3. Retry with Exponential Backoff and Jitter
7.2.4. Timeouts: Why They Are So Hard to Configure
7.2.5. Fallback Strategies: Degraded Service, Cached Response
7.2.6. From Hystrix to Resilience4j: Library Comparison
7.3. Failure Detection and Membership Protocols
Chapter Overview
7.3.1. Heartbeats and the Science of Timeout Configuration
7.3.2. Gossip Protocols: Information Dissemination via Epidemic Algorithms
7.3.3. Phi Accrual Failure Detector: Viewing Failure as a Probability
7.3.4. SWIM Protocol: Scalable Membership
7.3.5. Liveness vs. Readiness: The Difference Between Kubernetes Probes
7.4. Chaos Engineering
Chapter Overview
7.4.1. Chaos Monkey: Netflix's Pioneering Approach
7.4.2. Principles of Chaos Engineering: Hypothesis, Blast Radius, Observation
7.4.3. Fault Injection: Simulating Latency, Exceptions, and Partitions
7.4.4. GameDay: Disaster Scenario Drills
7.4.5. Automating Chaos: Litmus, Chaos Toolkit
VIII. Observability
Part Overview
8.1. The Three Pillars: Logs, Metrics, Traces
Chapter Overview
8.1.1. Structured Logging: JSON Logs and Correlation IDs
8.1.2. Log Aggregation: ELK Stack (Elasticsearch, Logstash, Kibana)
8.1.3. Metrics: Counter, Gauge, Histogram, Summary
8.1.4. The RED Method (Rate, Errors, Duration) vs. the USE Method
8.1.5. Distributed Tracing: Span, Trace, and Baggage Propagation
8.1.6. Correlation: Joining Three Signals from a Single Failure Event
8.2. Tools and Standards
Chapter Overview
8.2.1. OpenTelemetry: The Vendor-Neutral Instrumentation Standard
8.2.2. Prometheus: Pull-Based Metric Collection and PromQL
8.2.3. Grafana: Dashboard Design and Alerting
8.2.4. Jaeger / Zipkin / Tempo: Distributed Tracing Backends
8.2.5. Loki: Log Aggregation with the Prometheus Philosophy
8.2.6. eBPF: Kernel-Level Observability (Cilium, Pixie)
8.3. SLOs, SLAs, and Error Budgets
Chapter Overview
8.3.1. SLI (Service Level Indicator): What Should Be Measured?
8.3.2. SLO (Service Level Objective): The Art of Setting Targets
8.3.3. SLA (Service Level Agreement): Legal Obligations
8.3.4. Error Budget: Balancing Innovation and Reliability
8.3.5. Toil: Measuring and Reducing Operational Burden
8.3.6. Practical Lessons from the Google SRE Book
IX. Container Orchestration & Kubernetes
Part Overview
9.1. Kubernetes: The Operating System for Distributed Systems
Chapter Overview
9.1.1. Pod, Deployment, StatefulSet, DaemonSet: When to Use Which
9.1.2. The Kubernetes Scheduler: The Mechanics of Placement Decisions
9.1.3. etcd: The Heart of Cluster State
9.1.4. Horizontal Pod Autoscaler (HPA) and KEDA: Event-Driven Scaling
9.1.5. Node Affinity, Taints and Tolerations: Fine-Grained Placement
9.1.6. The Operator Pattern: Extending Kubernetes
9.2. Container Internals & Kubernetes in Production
Chapter Overview
9.2.1. Container Internals: Namespaces, cgroups v2, and the OCI Runtime
9.2.2. Resource Management: Requests, Limits, QoS, OOMKill, and CPU Throttling
9.2.3. Kubernetes Networking: Services, kube-proxy, and CNI
9.2.4. Persistent Storage: PV, PVC, and the CSI Interface
9.2.5. Graceful Termination and PodDisruptionBudgets
9.2.6. Cluster Autoscaling: Cluster Autoscaler, Karpenter, and VPA
X. Service Mesh & Network Security
Part Overview
10.1. Service Mesh Architecture
Chapter Overview
10.1.1. Why Service Mesh Exists: L4 vs. L7 Problems
10.1.2. Envoy Proxy: The Backbone of the Data Plane
10.1.3. Istio: istiod and the Consolidated Control Plane
10.1.4. Linkerd: A Lightweight Alternative with a Rust-Based Proxy
10.1.5. Traffic Management: Load Balancing, Retry, and Fault Injection
10.1.6. Observability at the Mesh Level: Automatic Telemetry
10.1.7. Sidecarless Mesh: Istio Ambient Mode and Cilium Service Mesh
10.2. Network Policies and mTLS
Chapter Overview
10.2.1. Kubernetes Network Policy: Pod-Level Firewall
10.2.2. mTLS: Mutual Authentication and Encryption
10.2.3. Certificate Lifecycle: Cert-Manager, SPIFFE / SPIRE
10.2.4. eBPF Networking: Kernel-Level CNI with Cilium
10.2.5. Ingress vs. Gateway API: Next-Generation Traffic Management
XI. Security Architecture
Part Overview
11.1. Authentication and Authorization
Chapter Overview
11.1.1. OAuth 2.0 & OIDC: Identity Management in a Distributed Environment
11.1.2. JWT: Signing, Verification, and Refresh Strategies
11.1.3. API Key vs. Service Account: Service Identity
11.1.4. RBAC vs. ABAC: Authorization Models
11.1.5. Workload Identity: Service Identity with SPIFFE / SPIRE
11.2. Zero Trust Architecture
Chapter Overview
11.2.1. Engineering the Never Trust, Always Verify Principle
11.2.2. Micro-Segmentation: The Least-Privilege Network Model
11.2.3. Identity-Aware Proxy (IAP)
11.2.4. The BeyondCorp Model: Google's Approach
11.2.5. Continuous Verification: Dynamic Security Decisions
11.2.6. Policy as Code: OPA, Gatekeeper, and Kyverno
11.3. Secrets Management and Encryption
Chapter Overview
11.3.1. HashiCorp Vault: Dynamic Secrets, Leases, and Rotation
11.3.2. Kubernetes Secrets: Why They Are Not Enough
11.3.3. Envelope Encryption: The Key Encryption Key (KEK) Model
11.3.4. Encryption in Transit vs. At Rest: Both Are Mandatory
11.3.5. Supply Chain Security: The SLSA Framework and Sigstore
XII. Deployment Strategies & Operations
Part Overview
12.1. Deployment Patterns
Chapter Overview
12.1.1. Blue/Green Deployment: Instant Cutover and Rollback
12.1.2. Canary Release: Progressive Traffic Shifting
12.1.3. Rolling Update: Staged Update Strategy
12.1.4. Feature Flags: Decoupling Deployment from Release
12.1.5. A/B Testing: Experimentation via Traffic Splitting
12.1.6. Shadow Deployment: Cloning Production Traffic
12.1.7. Zero-Downtime Database Migrations: The Expand/Contract Pattern
12.2. GitOps and CI/CD
Chapter Overview
12.2.1. GitOps Principles: Git as the Single Source of Truth
12.2.2. ArgoCD and Flux: GitOps Operators for Kubernetes
12.2.3. Progressive Delivery: Automated Canary Releases with Flagger
12.2.4. Pipeline as Code: Reproducible CI/CD
12.2.5. Artifact Management: OCI Registry and Helm Charts
12.2.6. Environment Promotion: Dev to Staging to Production
12.3. Platform Engineering
Chapter Overview
12.3.1. Internal Developer Platform (IDP): What It Is and Why It Is Needed
12.3.2. Backstage: Service Catalog and Developer Portal
12.3.3. Golden Paths: Speed and Security Through Standardized Routes
12.3.4. Infrastructure as Code: Terraform and Pulumi
12.3.5. Cost Optimization: Monitoring and Optimizing Resource Utilization
12.4. Testing in Distributed Systems
Chapter Overview
12.4.1. Integration Testing: Service Boundaries and Contract Verification
12.4.2. End-to-End Testing: Balancing Coverage and Flakiness
12.4.3. Testing Event-Driven Systems: Async Flows and Eventual Consistency
12.4.4. Staging Fidelity: How Realistic Is Your Test Environment?
12.4.5. Test Data Management: Synthetic Data, Anonymization, and Seeding
12.4.6. Load and Performance Testing: k6, Locust, and Capacity Validation
12.4.7. Correctness Testing: Jepsen and Deterministic Simulation
XIII. Case Studies
Part Overview
13.1. Messaging Platform: Apache Kafka
Chapter Overview
13.1.1. Disk-Based Persistence and Zero-Copy Optimization
13.1.2. KRaft Mode: Eliminating the ZooKeeper Dependency
13.1.3. Consumer Group Rebalancing: Failure Scenarios
13.1.4. Exactly-Once Semantics: Idempotent Producer + Transactional API
13.1.5. Kafka Streams vs. ksqlDB: When to Use Which
13.2. Distributed Key-Value Store: DynamoDB / Cassandra
Chapter Overview
13.2.1. SLA-Driven Design: The P99 Latency Guarantee
13.2.2. DynamoDB's Data Model: Partition Key + Sort Key
13.2.3. Cassandra's Consistent Hashing Ring
13.2.4. Vector Clocks in the Dynamo Paper: Conflict Resolution and Its Legacy
13.2.5. The Hot Partition Problem: A Real-World Case Study
13.3. Microservices at Scale: Netflix, Uber, Airbnb
Chapter Overview
13.3.1. Netflix: The Birth of Chaos Engineering
13.3.2. Uber's Domain-Oriented Microservice Architecture (DOMA)
13.3.3. Airbnb's Journey from Monolith to Microservices
13.3.4. Postmortem Culture: The Blameless Retrospective
13.3.5. Technical Debt Management: Sustainability at Scale
13.4. Real-Time Analytics Platforms
Chapter Overview
13.4.1. Stream Join Challenges: Temporal Join Strategies
13.4.2. Apache Pinot / Druid: OLAP at Low Latency
13.4.3. From Lambda to Data Mesh: The Evolution of Data Architecture
13.4.4. Data Mesh Principles: Domain-Oriented Data Ownership
13.4.5. ML Serving at Scale: Feature Store and Model Registry
13.5. Distributed SQL: Spanner & CockroachDB
Chapter Overview
13.5.1. Spanner: TrueTime, Paxos Groups, and External Consistency
13.5.2. CockroachDB: Raft, HLC, and Parallel Commits
13.5.3. Distributed Transactions in Practice: 2PC over Consensus
13.5.4. Follower Reads and Bounded Staleness
13.6. Anatomy of Famous Outages
Chapter Overview
13.6.1. GitHub 2018: 43 Seconds of Partition, 24 Hours of Recovery
13.6.2. Roblox 2021: The 73-Hour Consul Outage
13.6.3. Cloudflare 2019: The Regex That Stalled the Edge
13.6.4. Meta 2021: BGP, DNS, and Control Plane Centralization
GitHub
Select language
English
Türkçe
13.4.5. ML Serving at Scale: Feature Store and Model Registry