Mantinomia

Pattern	Short explanation	Example scenario where it should be used
Abstract Factory	Creates families of related objects without specifying their exact classes.	A desktop app needs matching Windows, macOS, and Linux UI widgets.
Blackboard	Lets components communicate through a shared central knowledge store.	A game AI system lets pathfinding, enemy detection, and strategy modules share world information.
Bridge / Bridges	Connects mismatched interfaces so components can communicate.	A new REST service must communicate with an older SOAP-based payment system.
Broker	Uses an intermediary between clients and service providers.	A distributed app uses a broker so clients do not need to know which server handles each request.
Client-Server	Splits the system into clients that request services and servers that provide them.	A mobile app asks a backend server for user profiles, messages, or files.
Composite	Treats individual objects and groups of objects through the same interface.	A GUI framework treats buttons, panels, and nested panels as drawable UI elements.
Dependency Injection	Provides an object’s dependencies from the outside instead of creating them internally.	A service receives a fake database during testing and a real database in production.
Entity Component System	Builds objects from flexible components instead of deep inheritance trees.	A game creates an enemy from Position, Health, AI, and Renderable components.
Factory Method	Lets a method decide which concrete object type to create.	A file importer creates a CSV parser, JSON parser, or XML parser depending on the file type.
Hierarchical Task Trees	Breaks complex behavior into a tree of smaller tasks.	A game NPC has a task tree for attacking the player: move closer, aim, shoot, and reload.
Intercepting Filter	Adds processing steps before or after a request reaches its target.	A web app applies logging, authentication, and compression filters around HTTP requests.
Intercepting Validator	Checks or validates a message before it reaches the target component.	An API gateway checks request signatures and JSON schema before forwarding traffic.
Layers / Layered	Organizes a system into levels where each layer uses the layer below it.	A business app separates presentation, business logic, and database access.
Map-Reduce	Splits large data processing into smaller parallel tasks and combines the results.	A log-analysis system counts events across terabytes of server logs.
Mediator / Mediators	Coordinates communication between components so they do not depend directly on each other.	Several business systems exchange orders through a mediator that translates and routes messages.
Memento	Saves and restores an object’s previous state without exposing its internals.	A text editor stores document snapshots so the user can undo changes.
Microservice Architecture	Builds a system from small, independently deployable services.	An e-commerce system separates catalog, cart, payment, and shipping into separate services.
Model-View-Controller / MVC	Separates data, user interface, and control logic.	A web app keeps product data in the model, HTML in the view, and user actions in the controller.
Monitor-Actuator	Separates monitoring logic from actuator execution so commands can be checked first.	A robot arm verifies that a movement command is safe before the motor performs it.
Multi-Tier	Splits a system into logical tiers, such as presentation, application, and data.	A banking system separates web frontend, application server, and database server.
Observer	Notifies dependent objects automatically when something changes.	A stock-price app updates several charts automatically when the price changes.
Peer-to-Peer	Lets components act as equals, both requesting and providing services.	A file-sharing app lets each user both download from and upload to other peers.
Pipe & Filter / Pipe-and-Filter	Sends data through a sequence of independent processing steps.	A compiler passes source code through lexing, parsing, optimization, and code generation.
Plug-in / Microkernel	Keeps a small core system and adds optional features through plug-ins.	An IDE lets users install plug-ins for Python, Java, Git, or debugging.
Publish-Subscribe	Publishers send events, and subscribers receive the events they are interested in.	A notification system sends “order shipped” events to email, SMS, and analytics subscribers.
Separated Safety	Keeps safety-critical parts separate from non-critical parts.	A train-control system keeps emergency braking logic separate from passenger information screens.
Service Mesh	Manages communication between microservices through a dedicated communication layer.	A microservice system uses a service mesh for routing, retries, encryption, and monitoring.
Service-Oriented Architecture / SOA	Builds systems from independent services that communicate through standard interfaces.	An enterprise system connects billing, inventory, and customer services across different platforms.
Shared-Data	Lets components communicate by reading and writing shared persistent data.	Several hospital systems read and write patient records through a shared database.
Singleton	Ensures that a class has only one instance.	A program uses one shared configuration object loaded at startup.
Strategy	Lets an object change its behavior by swapping the algorithm it uses.	A payment system chooses different fee-calculation strategies for credit card, invoice, or PayPal.
Template Method	Defines the overall structure of an algorithm while subclasses fill in specific steps.	A report generator always loads data, formats it, and exports it, but each report type customizes the steps.
Wrapper / Wrappers	Surrounds a component to adapt, hide, or preserve its interface.	A new system wraps an old legacy module so it can be used without changing its code.

Category	Tactic / mechanism	Why / when to use it	Example scenario	Relevant quality attributes, ranked
Detect faults	Monitor	Tracks component health so faults or congestion are detected early.	A service monitor notices that a backend server stopped responding.	Availability > Performance > Security
Detect faults	Ping/Echo	Sends a message and waits for a reply to check reachability and latency.	A load balancer pings servers before routing traffic to them.	Availability > Performance
Detect faults	Heartbeat	Sends periodic “I am alive” messages. Useful for detecting failed processes quickly.	A primary service expects heartbeats from a replica every few seconds.	Availability
Detect faults; Unsafe State Detection	Timestamp	Uses time/order information to detect incorrect event ordering.	A distributed system rejects an old sensor message.	Safety > Availability
Detect faults; Unsafe State Detection	Condition Monitoring	Checks process, device, or data state to detect faults.	A system verifies checksums to detect corruption.	Availability > Safety
Detect faults; Unsafe State Detection	Sanity Checking	Checks whether values or behavior stay within plausible limits.	A controller rejects a temperature reading of 10,000°C.	Safety > Availability
Detect faults	Voting	Compares outputs from replicas and chooses the majority or detects disagreement.	Three processors compute a value and the system accepts the majority result.	Availability > Safety
Detect faults	Exception Detection	Identifies deviations from expected behavior.	A service detects that a request handler entered an invalid state.	Availability > Testability
Detect faults	Self-Test	Lets a component check its own operation.	A device runs diagnostic checks during startup.	Availability > Testability
Recover from faults / Preparation and Repair	Redundant Spare	Keeps a backup component ready to take over.	A backup database server takes over when the primary fails.	Availability
Recover from faults / Preparation and Repair; Manage Deployment Pipeline; Recovery	Rollback	Restores a previous safe state or previous software version.	A failed deployment is rolled back to the last stable version.	Availability > Deployability > Safety
Recover from faults / Preparation and Repair	Exception Handling	Catches and manages runtime errors instead of letting the system crash.	A payment service catches a timeout and returns a controlled error.	Availability > Modifiability
Recover from faults / Software Upgrade	Functional Path	Loads updated code into pre-allocated RAM at runtime.	A control system loads a patched function while operating.	Deployability > Availability
Recover from faults / Software Upgrade	Class Patch	Adds or changes class data/methods at runtime.	A running application patches a class to fix a critical bug.	Deployability > Modifiability > Availability
Recover from faults / Software Upgrade	Hitless Upgrade	Uses redundancy so upgrades happen without visible interruption.	One server instance is upgraded while another keeps serving users.	Deployability > Availability
Recover from faults / Reintroduction	Shadow	Runs a failed or upgraded component in shadow mode before trusting it.	A new recommendation service receives real traffic but its output is not used yet.	Availability > Deployability > Testability
Recover from faults / Reintroduction	State Resynchronization	Aligns state between active and standby components.	A standby database receives checkpoints from the active database.	Availability
Recover from faults / Reintroduction	Escalating Restart	Restarts increasingly larger parts of the system.	Restart one service, then a subsystem, then the whole node if needed.	Availability
Recover from faults / Reintroduction	Non-Stop Forwarding	Keeps data flowing through alternate paths during failure.	A router keeps forwarding packets while its control plane restarts.	Availability > Performance
Recover from faults / Preparation and Repair	Retry	Reattempts failed operations. Useful when faults are temporary.	A service retries a failed network request after a short delay.	Availability > Performance
Recover from faults / Preparation and Repair	Ignore Faults / Ignore Faulty Behavior	Bypasses minor faults to keep the main service running.	A logging failure is ignored so the user request still completes.	Availability
Recover from faults / Preparation and Repair	Graceful Degradation	Continues with reduced capability instead of failing completely.	A video service lowers quality during overload.	Availability > Usability > Performance
Recover from faults / Preparation and Repair; Recovery	Reconfiguration	Changes component/resource assignments to maintain service or safety.	Work is moved from a failed server to a healthy one.	Availability > Safety
Prevent faults	Removal from Service	Takes faulty components offline to protect the rest of the system.	A server with high error rate is removed from the pool.	Availability > Performance
Prevent faults	Transactions	Ensures operations are atomic: all-or-nothing.	A bank transfer updates both accounts or neither.	Availability > Security
Prevent faults; Unsafe State Avoidance	Predictive Model	Predicts future faults or unsafe states from health/resource data.	A disk-health model warns before storage failure.	Safety > Availability
Prevent faults	Exception Prevention	Prevents faults using safer abstractions.	Smart pointers prevent invalid memory access.	Availability > Safety > Modifiability
Prevent faults	Increase Competence Set	Expands the situations the system can handle safely.	An autonomous vehicle is improved to handle more road conditions.	Availability > Safety
Manage Deployment Pipeline	Scaled Rollout	Gradually releases a new version to reduce deployment risk.	A new version is released to 5% of users first.	Deployability > Availability > Testability
Manage Deployment Pipeline	Scripted Deployment	Automates deployment with documented, tested scripts.	A CI/CD pipeline deploys using version-controlled scripts.	Deployability > Testability
Manage Deployed System	Service Interaction Management	Lets several service versions run at the same time.	API v1 and API v2 run during a migration period.	Deployability > Modifiability
Manage Deployed System	Dependency Packaging	Bundles correct dependency versions with the software.	A container image includes the exact libraries the service needs.	Deployability > Modifiability
Manage Deployed System	Feature Toggles	Turns features on/off at runtime.	A broken feature is disabled using a kill switch.	Deployability > Modifiability > Availability
Monitor Resources	Metering	Measures real energy use.	A server tracks power draw per workload.	Energy Efficiency
Monitor Resources	Static Classification	Estimates energy use from hardware/resource characteristics.	A system estimates power use from CPU and display type.	Energy Efficiency
Monitor Resources	Dynamic Classification	Estimates energy use from runtime workload data.	A phone predicts battery drain from current app behavior.	Energy Efficiency > Performance
Allocate Resources	Reduce Usage	Lowers energy use by tuning settings.	A phone lowers brightness and CPU frequency.	Energy Efficiency
Allocate Resources; Adapt; Defer Binding / Runtime binding	Discovery / Discover / Dynamic Discovery	Finds services/resources at runtime. Useful when available providers change.	A service registry tells a client where active service instances are.	Integrability > Modifiability > Deployability > Energy Efficiency
Allocate Resources; Manage Resources	Resource Scheduling / Schedule Resources	Allocates resources based on contention, energy, cost, or performance.	A scheduler gives CPU time to high-priority tasks first.	Performance > Energy Efficiency
Reduce Resource Demand; Control Resource Demand	Manage Event Arrival / Event Management	Controls how often events enter the system.	A sensor reduces sampling frequency to save battery.	Energy Efficiency > Performance
Reduce Resource Demand; Control Resource Demand	Limit Event Response	Queues, drops, or limits responses when capacity is exceeded.	A system drops low-priority telemetry during overload.	Performance > Energy Efficiency > Availability
Reduce Resource Demand; Control Resource Demand	Prioritize Events	Handles important events before less important ones.	Emergency alarms are processed before normal status updates.	Performance > Safety > Energy Efficiency
Reduce Resource Demand; Control Resource Demand	Reduce Computational Overhead	Reduces the work required per event.	A repeated expensive calculation is cached or simplified.	Performance > Energy Efficiency
Reduce Resource Demand; Control Resource Demand	Bound Execution Times	Limits how long tasks may run.	A request is cancelled if it exceeds 200 ms.	Performance > Safety > Energy Efficiency
Reduce Resource Demand; Control Resource Demand	Increase Resource Efficiency / Increase Resource Usage Efficiency	Improves algorithms or resource use so less work is needed.	Replace an O(n²) algorithm with an O(n log n) algorithm.	Performance > Energy Efficiency
Reduce Resource Demand	Workload Reduction	Reduces total system work.	A system skips non-essential background synchronization.	Energy Efficiency > Performance
Limit Dependencies; Reduce Coupling	Encapsulate	Hides internals behind a clear interface. Nice because changes do not spread.	A database module hides SQL details from business logic.	Modifiability > Integrability > Testability
Limit Dependencies; Reduce Coupling	Use an Intermediary	Inserts a mediator/facade/broker-like element to reduce direct dependencies.	A message broker connects services that should not call each other directly.	Integrability > Modifiability
Limit Dependencies; Reduce Coupling	Restrict Communication Paths	Limits which components may communicate.	Only approved services may call the payment service.	Integrability > Security
Limit Dependencies	Adhere to Standards	Uses shared standards to make integration easier.	Systems exchange data using JSON over HTTP.	Integrability > Modifiability
Limit Dependencies; Reduce Coupling	Abstract Common Services	Unifies similar services behind one common interface.	Several logging systems are hidden behind one logging API.	Integrability > Modifiability
Adapt	Tailor Interface	Adapts an interface through translation, buffering, or smoothing.	An adapter converts feet to meters between two systems.	Integrability > Modifiability
Adapt	Configure Behavior	Adjusts component behavior at build or runtime.	A component is configured to use API v1 or API v2.	Integrability > Modifiability
Coordinate	Orchestrate	Coordinates independent services through central control logic.	A workflow engine coordinates order, payment, and shipping.	Integrability > Modifiability
Coordinate	Manage Resources	Controls access to shared resources such as threads, memory, or devices.	A resource manager decides which service may use limited threads.	Integrability > Performance
Increase Cohesion	Split Module	Divides a large module into smaller, more focused modules.	A huge UserManager is split into authentication and profile modules.	Modifiability > Testability
Increase Cohesion	Redistribute Responsibilities	Moves responsibilities so related behavior is grouped better.	Validation logic is moved from controllers into a validation module.	Modifiability > Testability
Reduce Coupling	Restrict Dependencies	Limits which modules may depend on each other.	UI code is forbidden from directly accessing database code.	Modifiability > Testability
Defer Binding / Compile-time binding	Component Replacement	Allows components to be swapped through build/deployment mechanisms.	A build script selects either a mock or production payment component.	Modifiability > Deployability
Defer Binding / Compile-time binding	Compile-Time Parameters	Configures values during compilation.	A product variant is compiled with region-specific settings.	Modifiability
Defer Binding / Compile-time binding	Aspects	Separates cross-cutting concerns from core logic.	Logging or security checks are applied across many classes using aspects.	Modifiability > Testability
Defer Binding / Deployment or startup binding	Configuration Binding	Sets behavior through configuration at deployment/startup.	A service endpoint is changed through a config file.	Modifiability > Deployability
Defer Binding / Deployment or startup binding	Resource Files	Externalizes parameters into files.	UI text is stored in language resource files.	Modifiability > Deployability
Defer Binding / Runtime binding	Parameter Interpretation	Controls execution through runtime variables.	A feature changes behavior based on runtime parameters.	Modifiability
Defer Binding / Runtime binding	Shared Repositories	Coordinates through shared data.	Components read shared configuration from a repository.	Modifiability > Integrability
Defer Binding / Runtime binding	Polymorphism	Uses one interface with multiple implementations.	A payment interface has card, invoice, and PayPal implementations.	Modifiability > Testability
Control Resource Demand / Manage work requests	Manage Work Requests	Limits incoming work before overload happens.	An API sets a maximum request arrival rate.	Performance > Availability
Manage Resources	Increase Resources	Adds faster or more CPUs, memory, or network capacity.	Add more memory to reduce swapping.	Performance > Availability
Manage Resources	Introduce Concurrency	Processes work in parallel.	Multiple requests are handled by worker threads.	Performance
Manage Resources	Maintain Multiple Copies of Computations	Replicates computations to reduce contention.	Several servers perform the same calculation for different users.	Performance > Availability
Manage Resources	Maintain Multiple Copies of Data	Keeps data copies at different locations/speeds.	Frequently used data is cached near the application.	Performance > Availability
Manage Resources	Bound Queue Sizes	Limits queued work to control latency and memory use.	A web server rejects requests when its queue is full.	Performance > Availability
Unsafe State Avoidance	Substitution	Uses protective mechanisms such as watchdogs, monitors, or interlocks.	A watchdog restarts a controller that stops responding.	Safety > Availability
Unsafe State Detection	Timeout	Detects operations that exceed timing constraints.	A robot stops if a movement command takes too long.	Safety > Availability > Performance
Unsafe State Detection	Comparison	Compares synchronized or replicated outputs.	Two sensors disagree, so the system enters safe mode.	Safety > Availability
Containment / Redundancy; Availability Patterns	Replication / Server Replication	Duplicates components, data, or servers.	Several ticket servers handle high user load and tolerate failure.	Availability > Performance > Safety
Containment / Redundancy	Analytic Redundancy	Derives the same value through different calculations or sources.	Speed is estimated from both wheel rotation and GPS.	Safety > Availability
Containment / Redundancy	Functional Redundancy	Uses alternative functions to achieve a safe result.	If automatic steering fails, emergency braking can still prevent collision.	Safety > Availability
Containment / Limit Consequences	Masking	Hides or compensates for faults, often using redundancy.	A bad sensor reading is outvoted by other sensors.	Safety > Availability
Containment / Limit Consequences	Abort	Stops an unsafe operation immediately.	A robot stops when a human enters a danger zone.	Safety
Containment / Limit Consequences	Degradation	Continues only critical functions.	A vehicle disables comfort features but keeps braking and steering.	Safety > Availability
Containment / Barrier; Resist Attacks / Limit Access	Firewall / Firewall Protection	Restricts access to resources or network paths.	Payment services are placed behind a firewall.	Security > Safety > Availability
Containment / Barrier	Interlock	Prevents dangerous event sequences.	A train door cannot open while the train is moving.	Safety
Recovery	Repair State	Fixes erroneous state and resumes execution.	A system corrects corrupted control variables before continuing.	Safety > Availability
Detect Attacks	Detect Intrusion	Detects suspicious behavior using known malicious patterns.	A system flags requests matching SQL injection signatures.	Security > Availability
Detect Attacks	Detect Service Denial	Detects denial-of-service traffic patterns.	A server detects an abnormal flood of requests.	Security > Availability > Performance
Detect Attacks	Verify Message Integrity	Uses hashes or checksums to detect tampering.	A system verifies that a configuration file was not changed.	Security
Detect Attacks	Detect Message Delivery Anomalies / Detect Message Delay	Detects suspicious message timing or delivery behavior.	A system flags messages delayed in an unexpected pattern.	Security > Availability
Resist Attacks	Identify Actors	Identifies the source of external input.	A service records which client sent each request.	Security
Resist Attacks	Authenticate Actors	Verifies that an actor is who they claim to be.	A user logs in with password and two-factor authentication.	Security
Resist Attacks	Authorize Actors	Checks whether an authenticated actor has permission.	A normal user is blocked from admin-only operations.	Security
Resist Attacks	Limit Access	Restricts access to resources or entry points.	Only internal services can access the database.	Security > Safety
Resist Attacks	Limit Exposure	Reduces the attack surface.	A service exposes only one public API endpoint.	Security
Resist Attacks	Encrypt Data / Secure Connections	Protects data and communication from unauthorized reading.	Credit card data is sent over HTTPS.	Security
Resist Attacks	Separate Entities	Isolates systems using networks, VMs, or air gaps.	Payment processing runs on a separate network segment.	Security > Safety
Resist Attacks	Validate Input	Sanitizes and checks incoming data.	A web form rejects SQL injection attempts.	Security > Availability
Resist Attacks	Change Credential Settings / Change Default Credentials	Forces unsafe default credentials to be changed.	A router requires the admin password to be changed on first login.	Security
React to Attacks	Revoke Access	Removes access when an attack or compromise is suspected.	A compromised API key is disabled.	Security
React to Attacks	Restrict Login	Limits login attempts after repeated failures.	An account is locked after too many wrong passwords.	Security > Availability
React to Attacks	Inform Actors	Alerts people or systems about suspicious activity.	The security team receives an alert after intrusion detection.	Security > Availability
Recover from Attacks	Audit	Records actions so attacks or misuse can be traced.	A system logs who accessed patient records.	Security > Testability
Recover from Attacks	Nonrepudiation	Prevents parties from denying that they sent or received something.	A signed transaction proves who approved a payment.	Security
Control and Observe System State	Specialized Interfaces	Adds interfaces for controlling or observing state during testing.	A test API exposes internal component state.	Testability
Control and Observe System State	Record / Playback	Records interaction data and replays it during testing.	A test replays production traffic to reproduce a bug.	Testability
Control and Observe System State	Localize State Storage / Centralized State	Stores state in one controlled place to simplify test setup.	Tests load all state from a test database.	Testability > Modifiability
Control and Observe System State	Abstract Data Sources	Allows real data sources to be replaced with test sources.	A fake sensor replaces a real sensor during testing.	Testability > Modifiability
Control and Observe System State	Sandbox / Sandboxing	Isolates tests from real consequences.	A payment system is tested without charging real cards.	Testability > Security
Control and Observe System State	Executable Assertions	Embeds checks that detect faulty states.	Code asserts that account balance cannot become negative.	Testability > Availability
Limit Complexity	Limit Structural Complexity	Reduces cycles and coupling.	A cyclic dependency between modules is removed.	Testability > Modifiability
Limit Complexity	Limit Non-Determinism	Reduces uncontrolled randomness or concurrency.	Tests use fixed random seeds and controlled thread scheduling.	Testability
Support User Initiative	Cancel	Lets users stop an operation and frees resources.	A user cancels a large file upload.	Usability > Performance
Support User Initiative	Undo	Restores earlier states after user mistakes.	A text editor undoes the last edit.	Usability > Safety
Support User Initiative	Pause / Resume	Temporarily stops and later continues work.	A download is paused and resumed later.	Usability > Availability
Support User Initiative	Aggregate	Groups objects for collective operations.	A user selects many emails and archives them together.	Usability
Support System Initiative	Maintain Task Model	Tracks the user’s current task context.	A wizard knows which step the user is on.	Usability
Support System Initiative	Maintain User Model	Tracks user knowledge or behavior to tailor responses.	A system gives more guidance to beginners than experts.	Usability
Support System Initiative	Maintain System Model	Keeps a model of system state to provide accurate feedback.	A dashboard shows whether synchronization is pending or complete.	Usability
Mobile Systems Tactics	Power Source Monitoring	Detects low energy and triggers battery-saving behavior.	A phone alerts apps that shutdown may happen soon.	Energy Efficiency > Availability > Usability
Mobile Systems Tactics	Throttling Energy Usage	Degrades or terminates high-energy functionality.	A phone lowers CPU speed and sensor frequency.	Energy Efficiency > Availability
Mobile Systems Tactics	Tolerating a Loss of Power	Handles power failure and restart safely.	A mobile app saves state before shutdown and resumes later.	Availability > Energy Efficiency
Network connectivity	Support Only Necessary Network Interfaces	Uses only needed network interfaces to save power and reduce heat.	A device disables Wi-Fi scanning when Bluetooth is enough.	Energy Efficiency > Security
Network connectivity	Seamless Protocol Transition	Maintains connectivity while moving between networks.	A phone switches from Wi-Fi to cellular without dropping a call.	Availability > Usability
Network connectivity	Dynamic Protocol Selection	Chooses protocol based on cost, bandwidth, and power.	A mobile app chooses Wi-Fi for large uploads and cellular for small messages.	Energy Efficiency > Performance > Availability
Availability Patterns	Active Redundancy / Hot Spare	Runs backup nodes in parallel with the active node.	Flight-control computers run the same calculation in parallel.	Availability > Safety
Availability Patterns	Passive Redundancy / Warm Spare	Keeps a standby partly updated.	A warm database replica takes over after the primary fails.	Availability
Availability Patterns	Spare / Cold Spare	Keeps backup inactive until needed.	A backup server is powered on only after the main server fails.	Availability > Energy Efficiency
Availability Patterns	Triple Modular Redundancy / TMR	Runs three copies and uses voting.	A spacecraft computer accepts the majority result from three processors.	Availability > Safety
Availability Patterns	Circuit Breaker	Stops repeated calls to a failing service.	An order service stops calling a failing payment gateway.	Availability > Performance
Availability Patterns	Process Pairs	Uses checkpointing and a paired spare process.	A transaction process fails and its pair resumes from the latest checkpoint.	Availability
Availability Patterns	Forward Error Recovery	Moves from an erroneous state to a safe/correct state without rollback.	A drone switches to safe-flight mode after detecting bad sensor data.	Availability > Safety
Deployability Patterns	Blue/Green Deployment	Runs old and new versions side by side, then switches traffic.	Traffic switches from blue production to green production.	Deployability > Availability
Deployability Patterns	Rolling Upgrade	Updates instances one at a time.	Kubernetes replaces pods gradually.	Deployability > Availability
Deployability Patterns	Canary Testing	Releases to a small user group first.	1% of users get the new search algorithm first.	Deployability > Testability > Availability
Deployability Patterns	A/B Testing	Compares alternatives with real users.	A shop tests two checkout flows.	Usability > Deployability
Energy Efficiency Patterns	Sensor Fusion	Uses low-power sensors to decide when high-power sensors are needed.	An app uses accelerometer data before activating GPS.	Energy Efficiency > Performance > Safety
Energy Efficiency Patterns	Kill Abnormal Tasks	Stops tasks with abnormal energy use.	A mobile OS stops an app constantly using GPS.	Energy Efficiency > Availability
Energy Efficiency Patterns	Power Monitor	Tracks power and disables unused devices/functions.	A phone disables unused sensors while idle.	Energy Efficiency
Performance Patterns	Load Balancer	Distributes requests across servers.	A web shop spreads traffic across many backend servers.	Performance > Availability
Performance Patterns	Throttling	Limits request rate or workload.	An API allows each user 100 requests per minute.	Performance > Availability > Security
Patterns for Safety	Redundant Sensors	Duplicates sensors for safer data.	A self-driving car compares several sensor readings before braking.	Safety > Availability
Patterns for Security	Intrusion Prevention System / IPS	Identifies, analyzes, blocks, and reports suspicious activity.	A network IPS blocks traffic matching known attack behavior.	Security > Availability
Security design concept	User Authentication / Authorization	Verifies identity and permissions.	Only logged-in users can buy tickets, and only admins can manage matches.	Security
Secure payment design concept	Third-Party Payment Processing	Moves sensitive payment handling to a trusted external service.	A webshop sends card payment to a certified payment provider.	Security > Modifiability
Modifiability design concept	Modular Design / High Cohesion and Low Coupling	Groups related responsibilities and reduces unnecessary dependencies.	External team-server integration is isolated from core ticket logic.	Modifiability > Integrability > Testability
Integrability design concept	Adaptable Interfaces for External Systems	Makes changing external systems easier to integrate.	A ticket system adapts to different football team servers.	Integrability > Modifiability
Performance design concept	Data Caching	Stores frequently used data closer to where it is needed.	Match and seat data are cached instead of fetched every time.	Performance > Availability
Modifiability design concept	Separation of Core Functionality and External Interfaces	Keeps stable logic separate from change-prone integration code.	Ticket booking logic is separate from team-server adapters.	Modifiability > Integrability
Security design concept	Separation of Secure Parts	Isolates security-sensitive functionality.	Payment logic is separated from normal browsing logic.	Security > Modifiability