Reliability Patterns

Reliability patterns aim to ensure systems function as expected, even when components fail. They focus on designing systems that can recover from errors, handle unexpected inputs, and maintain a consistent level of service.

Wikipedia

Load Shedding

reliability resilience behavioral

Load shedding is a mechanism for gracefully handling overload situations in a system. When demand exceeds available resources (CPU, memory, network, database connections, etc.), a load shedding strategy is employed to selectively reject or delay requests, preventing the system from complete failure. It prioritizes …

Read More
Outbox Pattern

behavioral integration reliability

The Outbox Pattern addresses the challenge of reliably publishing events in a transactional context. When an application modifies its data and needs to publish an event to notify other services, ensuring both operations occur atomically is crucial. Directly publishing events from the application can lead to eventual …

Read More
Transactional Consumer

behavioral integration reliability

The Transactional Consumer pattern ensures that messages received from a message broker are processed reliably and atomically. It handles message consumption within a database transaction, allowing for either complete processing of a message batch or a full rollback in case of failure. This prevents partial updates and …

Read More
Transactional Outbox

behavioral integration reliability

The Transactional Outbox pattern solves the problem of reliably publishing events in a microservices architecture when an event needs to be triggered as a direct result of a database transaction. It ensures that events are not published if the transaction fails, maintaining data consistency. Instead of directly …

Read More
Shadow Deployment

behavioral deployment reliability

Shadow Deployment is a technique to test changes in a live production environment without impacting real users. It involves duplicating real production traffic to a new, “shadow” version of the application. This allows for observing the new system’s behavior under realistic load, identifying potential …

Read More
Canary Release

behavioral deployment reliability

The Canary Release pattern is a deployment strategy where a new version of software is rolled out to a small subset of users or servers before being made available to the entire infrastructure. This allows for real-world testing of the new version with minimal impact, enabling teams to detect and address any unforeseen …

Read More
Geo-Replication

distributed systems scalability reliability

Geo-Replication is a technique used to distribute data across multiple geographically diverse locations. This is done to improve performance for users in those regions (by reducing latency), increase availability and fault tolerance (by having backups in different locations), and provide disaster recovery capabilities. …

Read More
Cluster-based Architecture

architectural scalability reliability

A Cluster-based Architecture involves grouping multiple interconnected computers (nodes) together to work as a single system. This approach enhances performance, availability, and scalability by distributing workloads across the cluster. The nodes typically share resources and are managed by software that coordinates …

Read More
Replication

distributed systems reliability scalability

The Replication pattern addresses the need for data consistency and availability across multiple systems. It involves creating and maintaining multiple copies of data, ensuring that if one copy fails, others are available to serve requests. This enhances fault tolerance, improves read performance by distributing load, …

Read More
Two-Phase Commit

distributed system concurrency reliability

The Two-Phase Commit (2PC) protocol is a distributed algorithm used to ensure atomic commits across multiple transaction participants. In essence, it guarantees that either all participants commit changes to the database or none do, maintaining data consistency in a distributed environment. This is crucially important …

Read More