Visão Geral
Este curso aborda o desenho de pipelines de dados de métricas utilizando Telegraf e InfluxDB, com foco em arquitetura, confiabilidade, escalabilidade e boas práticas. O aluno aprenderá a projetar fluxos completos de coleta, processamento, envio e armazenamento de dados de séries temporais para ambientes corporativos, industriais e de observabilidade.
Conteúdo Programatico
Module 1 – Data Pipelines and Observability Concepts
- What is a data pipeline
- Metrics pipeline fundamentals
- Observability architecture overview
- Common pipeline anti-patterns
Module 2 – Telegraf as a Data Pipeline Agent
- Telegraf internal data flow
- Inputs, processors and outputs roles
- Plugin chaining strategy
- Pipeline design considerations
Module 3 – InfluxDB as Time Series Storage
- InfluxDB ingestion model
- Buckets and retention design
- Write performance characteristics
- Storage optimization
Module 4 – Pipeline Architecture Design
- Centralized vs distributed collection
- Edge vs core processing
- High availability architectures
- Secure pipeline design
Module 5 – Data Modeling Strategy
- Measurements, fields and tags design
- Cardinality impact on pipelines
- Naming standards
- Schema evolution
Module 6 – Reliability and Fault Tolerance
- Buffering and batching strategies
- Retry mechanisms
- Handling network failures
- Data loss prevention
Module 7 – Performance and Scalability
- Load distribution
- Scaling Telegraf agents
- InfluxDB ingestion tuning
- Capacity planning
Module 8 – Real-World Pipeline Use Cases
- Infrastructure monitoring pipelines
- Application metrics pipelines
- Industrial and IoT data pipelines
- Best practices and lessons learned