ITEC 4010 - L5: Reliability, Scalability, and Maintainability

Data-Intensive Application

A system whose primary challenge is managing, storing, and processing large volumes of data efficiently rather than computation speed.

Reliability

The ability of a system to continue functioning correctly even when faults occur.

Fault

A condition where a component deviates from its specification or intended behavior.

Failure

A situation where the entire system stops providing correct service or expected output.

Fault-Tolerant System

A system designed to continue operating properly even in the presence of faults.

Resilience

The capability of a system to recover and continue functioning after experiencing faults.

Hardware Faults

Failures caused by physical components such as disks, memory, power supplies, or network cables.

Redundancy

The use of duplicate components or systems to prevent single points of failure from causing outages.

Software Errors

Failures caused by bugs or unexpected behavior in code, often systematic and correlated across systems.

Human Errors

Mistakes made by operators, developers, or administrators that lead to system malfunctions.

Fault Injection Testing

A deliberate practice of introducing controlled failures to verify system reliability and resilience (e.g., Netflix Chaos Monkey).

Scalability

The capability of a system to handle increasing load or growth without degrading performance.

Load Parameter

A measurable characteristic representing system load, such as requests per second, active users, or database transactions.

Throughput

The rate at which a system completes tasks or processes data, often measured as operations per second.

Response Time

The total time between sending a request and receiving a response from the system.

Vertical Scaling

The process of improving system performance by upgrading hardware to a more powerful machine.

Horizontal Scaling

The process of increasing capacity by adding more machines or servers to distribute the workload.

Elastic Scaling

An automated method of dynamically adding or removing computing resources based on demand.

Fan-Out Problem

A scalability challenge where an operation must interact with many related users or systems, causing increased load (e.g., Twitter home timeline).

Maintainability

The ease with which engineers can understand, modify, fix, and operate a system over its lifetime.

Operability

A system design quality that ensures smooth and efficient operation through monitoring, automation, and clear documentation.

Simplicity

A design principle aimed at reducing accidental complexity and creating cleaner, more understandable systems.

Accidental Complexity

Unnecessary complexity that arises from poor design, unclear abstractions, or tangled dependencies.

Abstraction

The technique of hiding low-level implementation details behind simplified, clear interfaces or APIs.

Evolvability

The ease with which a system can be changed, adapted, or extended as requirements evolve.

Agile Methodology

An iterative development approach that supports evolvability through continuous improvement and flexibility.

Legacy Maintenance

The ongoing process of maintaining, updating, and improving existing systems while minimizing disruption.

Non-Functional Requirements

System qualities like reliability, scalability, and maintainability that define how a system performs rather than what it does.

Reliability Principle

The idea that faults are inevitable, so systems must be designed to prevent faults from turning into failures.

Redundancy Principle

A reliability approach that uses duplicate components to ensure system continuity during failures.

Scalability Principle

The concept that system design should allow proportional performance growth as load increases.

Maintainability Principle

A design focus that emphasizes simplicity, abstraction, and operability to make long-term changes easier.

Chaos Engineering

A discipline of intentionally injecting faults into production-like systems to validate reliability and fault tolerance.