登录查看更多内容

How do you document performance issues?

由人工智能和领英社区提供技术支持

Performance issues can affect the quality, reliability, and user experience of any software project. Whether you are developing a web app, a mobile game, or a desktop tool, you need to document how your code performs under different scenarios and conditions. Documenting performance issues can help you identify and resolve bottleneecks, optimize resource usage, and improve customer satisfaction. In this article, you will learn how to document performance issues using some common methods and tools.

此文章中的业界达人

由社区从 8 条内容中精选。了解更多

Krishna Roy

Senior Professional I at Capgemini
Otávio Celestino dos Santos

Senior Software Engineer @ Mercado Libre | Java / Kotlin / GoLang | 5x AWS & Terraform Certified, GoLang Tutor

1 Define performance criteria

The first step to document performance issues is to define what performance means for your project. Performance criteria are the measurable aspects of your software that affect its functionality and usability. For example, you might want to document the response time, throughput, scalability, availability, or reliability of your software. Performance criteria should be aligned with your project goals, user expectations, and industry standards. You should also specify the acceptable ranges, thresholds, or targets for each criterion.

添加您的观点

Krishna Roy

Senior Professional I at Capgemini
举报内容
A performance issue refers to any degradation or inefficiency within a system, application, or process that negatively impacts its speed, responsiveness, throughput, or overall efficiency. Common performance bottlenecks can be - Performance problems like Poorly optimized algorithms, slow code , CPU usage or memory inefficiencies. Concurrency issues such as deadlocks or thread problems. Memory leaks like high memory consumption. Exceptions and errors in the code. Data Retrieval and Processing like Fetching and processing large amounts of data extensively. Garbage collection inefficiencies affecting responsiveness. DB Optimization like Inefficient database queries. Transaction Management like improper transaction handling.

已翻译

赞
Otávio Celestino dos Santos

Senior Software Engineer @ Mercado Libre | Java / Kotlin / GoLang | 5x AWS & Terraform Certified, GoLang Tutor
举报内容
Definir critérios de desempenho é como estabelecer regras para garantir que seu software funcione da melhor maneira possível: - Entender o Significado de Desempenho:? Defina o que desempenho significa para o seu projeto e por quê ele deve ser a prioridade durante o desenvolvimento e testes. - Aspectos que s?o possíveis de “mensurar”: Identifique os aspectos mensuráveis que afetam o software, como erros no projeto, pontos de falha, etc. - Alinhamento com Metas do Projeto: Garanta que os critérios de desempenho estejam alinhados com as metas específicas do projeto ou time. - Especificar Intervalos e Limites: Estabele?a intervalos, limites ou metas aceitáveis como tempo de resposta esperado, requisi??es por segundo, e outras métricas.

已翻译

赞
Samuel Espinoza

Software Engineer | Tech Lead | 8+ years with Node.js, TypeScript, RESTful Microservices, Docker
举报内容
Performance criteria can be hard requirements depending on your business and the functionality that you're implementing. If you need a particular API endpoint to respond fast (< 1 sec) then a stress test should be in place to check the performance criteria. But if you have a feature that only needs to send an email, you don't really care when in the future that needs to happen. Choose wisely what truly needs to be tested for performance and focus on that.

已翻译

赞

2 Collect performance data

The next step is to collect performance data for your software in various situations and environments. You can use profiling tools, such as Visual Studio or Xcode, to analyze the behavior and resource consumption of your code during execution. This can help you pinpoint performance issues in your code. Testing tools, like JMeter or Gatling, can be used to simulate and measure the performance of your software under different loads or concurrency levels. This will help you evaluate the performance of your software against your criteria. Lastly, monitoring tools like Prometheus or New Relic can be used to track and visualize performance metrics, trends, and anomalies of your software in real-time or over a period of time. This can help you detect and diagnose any issues in your production environment.

添加您的观点

Krishna Roy

Senior Professional I at Capgemini
举报内容
Before collect the performance data, need to understand performance issue. Problem statement should be very clear and more descriptive that is causing the performance degradation in application Performance Data : JFR & Async Profiler. Heap dump. Thread Dump. Particular Logs. Centralized / Distributed logs.

已翻译

赞
Otávio Celestino dos Santos

Senior Software Engineer @ Mercado Libre | Java / Kotlin / GoLang | 5x AWS & Terraform Certified, GoLang Tutor
举报内容
Coletar dados de desempenho é como colocar seu software sob um microscópio: - Ferramentas de Perfil (Go Profiler, cProfile, Valgrind): Use ferramentas de cria??o de perfil para analisar como seu código se comporta e consome recursos durante a execu??o. - Ferramentas de Teste (JMeter, Gatling, GoLang Benchmarks): Simule diferentes cargas e níveis de simultaneidade com ferramentas de teste para medir como seu software performa em situa??es diversas. - Avalia??o em Rela??o aos Critérios: Compare os resultados com seus critérios de desempenho, avaliando se o software atende às metas estabelecidas. - Ferramentas de Monitoramento (Prometheus, New Relic, Grafana, Datadog): Utilize ferramentas de monitoramento para rastrear métricas em tempo real.

已翻译

赞

3 Analyze performance data

The third step is to analyze the data collected and identify performance issues that need to be addressed. You can use various methods and tools to analyze performance data, such as benchmarking, profiling, and reporting. Benchmarking involves comparing the performance of your software with other software or standards, using tools such as Apache Bench, Siege, or Wrk. Profiling interprets profiling data to find the root causes of performance issues, with tools like Visual Studio, Eclipse, or Xcode. Reporting summarizes and presents the performance data collected and analyzed, with tools like Excel, PowerPoint, or Google Docs. All of these techniques can help you understand the strengths and weaknesses of your software, improve its efficiency and quality, and communicate its performance results, issues, and recommendations.

添加您的观点

Krishna Roy

Senior Professional I at Capgemini
举报内容
To analyse the performance data you can use various methods and tools, few are open source and few are paid tools. JMC tool is used to analysed the JFR and Async data. Memory Analyzer tool for heap dump. Jaeger for Open Tracing. JStack / JMC foe Thread Dump. E.t.c...

已翻译

赞
Otávio Celestino dos Santos

Senior Software Engineer @ Mercado Libre | Java / Kotlin / GoLang | 5x AWS & Terraform Certified, GoLang Tutor
举报内容
Existem várias ferramentas para nos ajudar nesse formato: - Benchmarking: 1. Vegeta: Uma ferramenta de benchmarking HTTP que pode gerar cargas e medir o desempenho de servidores web. 2. hey: Uma ferramenta de benchmarking rápida e flexível para servi?os web HTTP/RESTful. - Cria??o de Perfil: 1. Go pprof: Uma ferramenta de perfiliza??o em Go, que permite analisar a aloca??o de memória e o tempo de CPU. 2. Java Mission Control (JMC): Uma ferramenta de cria??o de perfil para aplica??es Java. - Relatórios: 1. Jupyter Notebooks: Uma plataforma interativa que permite criar documentos contendo código e visualiza??es deal para análise e relatórios. 2. Tableau: Uma ferramenta de visualiza??o de dados que facilita a cria??o de relatórios.

已翻译

赞

4 Document performance issues

The final step is to document the performance issues that you identified and analyzed in a clear, concise, and consistent way. Various methods and tools can be used for this purpose, such as comments to explain or describe code performance, logs to report software performance, and documentation to describe the performance of your software. Comments can help you and other developers understand and maintain your code, logs can assist in troubleshooting and debugging your software, and documentation can help showcase and improve your software to stakeholders, clients, or users.

添加您的观点

Otávio Celestino dos Santos

Senior Software Engineer @ Mercado Libre | Java / Kotlin / GoLang | 5x AWS & Terraform Certified, GoLang Tutor
举报内容
Documentar os problemas de desempenho é a última pe?a: - Comentários no Código: * O que faz: Explica??es dentro do código para descrever decis?es de design, otimiza??es ou áreas de aten??o. * Ferramentas: Comentários no código-fonte usando linguagem específica. - Logs: * O que faz: Registra informa??es relevantes durante a execu??o do software, úteis para solu??o de problemas e depura??o. * Ferramentas: Log4j, Logback, Winston (para Node.js). - Documenta??o: * O que faz: Descreve o desempenho do software e suas otimiza??es para desenvolvedores, partes interessadas e usuários. * Ferramentas: Markdown, reStructuredText, ferramentas de gera??o de documenta??o como GoDoc ou Javadoc.

已翻译

赞

5 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?

添加您的观点

Programming

+ 关注

给文章评分

我们借助人工智能创建了此文章。您认为这篇文章怎么样？

很棒不太好

举报此文章

查看全部

How do you document performance issues?

1

2

3

4

5

1 Define performance criteria

2 Collect performance data

3 Analyze performance data

4 Document performance issues

5 Here’s what else to consider

Programming

给文章评分

感谢您的反馈

更多Programming相关文章

更多相关阅读内容

How do you document performance issues?

1

2

3

4

5

1 Define performance criteria

2 Collect performance data

3 Analyze performance data

4 Document performance issues

5 Here’s what else to consider

Programming

给文章评分

感谢您的反馈

查看其他技能