- 理解系统监控和可观察性概念
- 有使用Grafana或Prometheus的经验
- 熟悉Python和基本的机器学习原理
受众
- 可观察性工程师
- 基础设施和DevOps团队
- 监控平台架构师和站点可靠性工程师(SREs)
Prometheus 和 Grafana 是现代基础设施中广泛采用的可观测性工具,而机器学习通过提供预测性和智能化的洞察,增强了这些工具,以自动化运维决策。
本次由讲师指导的培训(线上或线下)面向中级可观测性专业人员,旨在通过整合 AIOps 实践,使用 Prometheus、Grafana 和机器学习技术,实现监控基础设施的现代化。
在培训结束时,参与者将能够:
- 配置 Prometheus 和 Grafana,以实现跨系统和服务的可观测性。
- 收集、存储和可视化高质量的时间序列数据。
- 应用机器学习模型进行异常检测和预测。
- 基于预测性洞察构建智能告警规则。
课程形式
- 互动式讲座和讨论。
- 大量练习和实践。
- 在实时实验室环境中进行实际操作。
课程定制选项
- 如需为此课程定制培训,请联系我们安排。
AIOps 开源工具介绍
- AIOps 概念与优势概述
- 可观测性栈中的 Prometheus 和 Grafana
- ML 在 AIOps 中的应用:预测性与反应性分析
Prometheus 和 Grafana 的配置
- 安装并配置 Prometheus 以收集时间序列数据
- 使用实时指标在 Grafana 中创建仪表板
- 探索导出器、重新标记和服务发现
ML 数据预处理
- 提取并转换 Prometheus 指标
- 准备用于异常检测和预测的数据集
- 使用 Grafana 的转换或 Python 管道
应用 Machine Learning 进行异常检测
- 异常检测的基本 ML 模型(如 Isolation Forest、One-Class SVM)
- 在时间序列数据上训练和评估模型
- 在 Grafana 仪表板中可视化异常
使用 ML 的 Forecasting 指标
- 构建简单的预测模型(ARIMA、Prophet、LSTM 简介)
- 预测系统负载或资源使用情况
- 使用预测结果进行早期警报和扩展决策
ML 与警报和自动化的集成
- 基于 ML 输出或阈值定义警报规则
- 使用 Alertmanager 和通知路由
- 在检测到异常时触发脚本或自动化工作流
AIOps 的扩展与操作化
- 集成外部可观测性工具(如 ELK stack、Moogsoft、Dynatrace)
- 在可观测性管道中操作化 ML 模型
- AIOps 在大规模应用中的最佳实践
总结与下一步
United Arab Emirates - Implementing AIOps with Prometheus, Grafana, and ML
Qatar - Implementing AIOps with Prometheus, Grafana, and ML
Egypt - Implementing AIOps with Prometheus, Grafana, and ML
Saudi Arabia - Implementing AIOps with Prometheus, Grafana, and ML
South Africa - Implementing AIOps with Prometheus, Grafana, and ML
Brasil - Implementing AIOps with Prometheus, Grafana, and ML
Canada - Implementing AIOps with Prometheus, Grafana, and ML
中国 - Implementing AIOps with Prometheus, Grafana, and ML
香港 - Implementing AIOps with Prometheus, Grafana, and ML
澳門 - Implementing AIOps with Prometheus, Grafana, and ML
台灣 - Implementing AIOps with Prometheus, Grafana, and ML
USA - Implementing AIOps with Prometheus, Grafana, and ML
Österreich - Implementing AIOps with Prometheus, Grafana, and ML
Schweiz - Implementing AIOps with Prometheus, Grafana, and ML
Deutschland - Implementing AIOps with Prometheus, Grafana, and ML
Czech Republic - Implementing AIOps with Prometheus, Grafana, and ML
Denmark - Implementing AIOps with Prometheus, Grafana, and ML
Estonia - Implementing AIOps with Prometheus, Grafana, and ML
Finland - Implementing AIOps with Prometheus, Grafana, and ML
Greece - Implementing AIOps with Prometheus, Grafana, and ML
Magyarország - Implementing AIOps with Prometheus, Grafana, and ML
Ireland - Implementing AIOps with Prometheus, Grafana, and ML
Luxembourg - Implementing AIOps with Prometheus, Grafana, and ML
Latvia - Implementing AIOps with Prometheus, Grafana, and ML
España - Implementing AIOps with Prometheus, Grafana, and ML
Italia - Implementing AIOps with Prometheus, Grafana, and ML
Lithuania - Implementing AIOps with Prometheus, Grafana, and ML
Nederland - Implementing AIOps with Prometheus, Grafana, and ML
Norway - Implementing AIOps with Prometheus, Grafana, and ML
Portugal - Implementing AIOps with Prometheus, Grafana, and ML
România - Implementing AIOps with Prometheus, Grafana, and ML
Sverige - Implementing AIOps with Prometheus, Grafana, and ML
Türkiye - Implementing AIOps with Prometheus, Grafana, and ML
Malta - Implementing AIOps with Prometheus, Grafana, and ML
Belgique - Implementing AIOps with Prometheus, Grafana, and ML
France - Implementing AIOps with Prometheus, Grafana, and ML
日本 - Implementing AIOps with Prometheus, Grafana, and ML
Australia - Implementing AIOps with Prometheus, Grafana, and ML
Malaysia - Implementing AIOps with Prometheus, Grafana, and ML
New Zealand - Implementing AIOps with Prometheus, Grafana, and ML
Philippines - Implementing AIOps with Prometheus, Grafana, and ML
Singapore - Implementing AIOps with Prometheus, Grafana, and ML
Thailand - Implementing AIOps with Prometheus, Grafana, and ML
Vietnam - Implementing AIOps with Prometheus, Grafana, and ML
India - Implementing AIOps with Prometheus, Grafana, and ML
Argentina - Implementing AIOps with Prometheus, Grafana, and ML
Chile - Implementing AIOps with Prometheus, Grafana, and ML
Costa Rica - Implementing AIOps with Prometheus, Grafana, and ML
Ecuador - Implementing AIOps with Prometheus, Grafana, and ML
Guatemala - Implementing AIOps with Prometheus, Grafana, and ML
Colombia - Implementing AIOps with Prometheus, Grafana, and ML
México - Implementing AIOps with Prometheus, Grafana, and ML
Panama - Implementing AIOps with Prometheus, Grafana, and ML
Peru - Implementing AIOps with Prometheus, Grafana, and ML
Uruguay - Implementing AIOps with Prometheus, Grafana, and ML
Venezuela - Implementing AIOps with Prometheus, Grafana, and ML
Polska - Implementing AIOps with Prometheus, Grafana, and ML
United Kingdom - Implementing AIOps with Prometheus, Grafana, and ML
South Korea - Implementing AIOps with Prometheus, Grafana, and ML
Pakistan - Implementing AIOps with Prometheus, Grafana, and ML
Sri Lanka - Implementing AIOps with Prometheus, Grafana, and ML
Bulgaria - Implementing AIOps with Prometheus, Grafana, and ML
Bolivia - Implementing AIOps with Prometheus, Grafana, and ML
Indonesia - Implementing AIOps with Prometheus, Grafana, and ML
Kazakhstan - Implementing AIOps with Prometheus, Grafana, and ML
Moldova - Implementing AIOps with Prometheus, Grafana, and ML
Morocco - Implementing AIOps with Prometheus, Grafana, and ML
Tunisia - Implementing AIOps with Prometheus, Grafana, and ML
Kuwait - Implementing AIOps with Prometheus, Grafana, and ML
Oman - Implementing AIOps with Prometheus, Grafana, and ML
Slovakia - Implementing AIOps with Prometheus, Grafana, and ML
Kenya - Implementing AIOps with Prometheus, Grafana, and ML
Nigeria - Implementing AIOps with Prometheus, Grafana, and ML
Botswana - Implementing AIOps with Prometheus, Grafana, and ML
Slovenia - Implementing AIOps with Prometheus, Grafana, and ML
Croatia - Implementing AIOps with Prometheus, Grafana, and ML
Serbia - Implementing AIOps with Prometheus, Grafana, and ML
Bhutan - Implementing AIOps with Prometheus, Grafana, and ML
Nepal - Implementing AIOps with Prometheus, Grafana, and ML
Uzbekistan - Implementing AIOps with Prometheus, Grafana, and ML