Performance Engineering

Reason for Topic

As modern, distributed software systems increase in inherent complexity, so to the challenges associated with coordinating and managing multiple components across a distributed environment. Aggregate systems made up of many microservices often make it more challenging to identify and diagnose performance and reliability issues, particularly when they arise from interactions between components.  

Being connected through distributed networks also results in unexpected latencies and failure modes, which can impact the overall performance and reliability of the system. A failure in one component can have cascading effects on other components in the system.  

Distributed systems often rely on shared resources, such as databases or network infrastructure. Resource contention can occur when multiple components compete for the same resource, leading to performance degradation and potential reliability issues. These systems also need to be designed and architected to scale to meet changing demands. However, improper scaling can lead to performance and reliability issues. 

For these and other reasons, performance engineering as a practice must be adopted throughout teams, technological landscapes, early in design and development cycles, and as part of deployment / transition phases of software delivery. 

Introduction / Definition

Performance engineering encompasses the techniques applied during a systems development life cycle to ensure the non-functional requirements for performance (such as throughput, latency, or memory usage) will be met. Performance engineering addresses issues in more than just the software and supporting infrastructure, and adherence to non-functional requirements is also validated post-deployment by monitoring the production systems. This is part of IT service management (see also ITIL). 

Benefits & Examples

Performance engineering helps to identify and eliminate performance bottlenecks early in the software development life cycle. By identifying these bottlenecks early, the development team can address them before they become major issues that impact the end-users. This results in a software application that is more reliable and performs better, leading to increased user satisfaction. 

Performance engineering helps to improve the scalability of a software application. Scalability is the ability of a software application to handle an increasing number of users or requests without a significant decrease in performance. Performance engineering ensures that the software application is designed and developed to handle increased traffic, ensuring that the application can scale to meet the demands of the users. 

Performance engineering helps to reduce the cost of maintaining costs associated with a software application. A well-performing software application requires less maintenance and fewer resources to maintain. By optimizing the performance of the software application, the development team can reduce the cost of maintaining the application over time. 

Performance engineering helps to improve the overall user experience of a software application. A software application that performs well and responds quickly to user requests leads to a better user experience. By optimizing the performance of the software application, the development team can ensure that the end-users have a positive experience when using the application. 

Performance engineering helps to improve the overall efficiency of the software development process. By identifying and addressing performance issues early in the development process, the development team can reduce the time and resources required to develop and test understand the characteristics of the application, produce accurate requirements, identify and fix defects, and more accurately anticipate resolutions to issues that arise in production.  

Drawbacks / Gotchas

• Complexity: Performance engineering requires specialized skills and knowledge to be done effectively. This can make the development process more complex and require additional resources. 
• Limited resources: Performance engineering can require significant computing resources and infrastructure to execute properly. Small teams or organizations with limited resources may find it difficult to invest in performance engineering. 
• Over-optimization: In some cases, performance engineering can lead to over-optimization, where the focus on performance comes at the expense of other critical aspects of software quality, such as usability or maintainability. 

• False positives: Performance testing tools may generate false positives or negatives, leading to incorrect conclusions about the performance of the software application. 

Summary

Performance engineering is a critical process that helps to improve the overall quality of a software application. By identifying and eliminating performance bottlenecks early in the development process, performance engineering helps to improve reliability, scalability, security, user experience, and efficiency, while also reducing the cost of maintaining the software application over time.