MLOps Academy · 课时

构建 Grafana 仪表板

一眼查看服务健康状况

构建 Grafana 仪表板 是 CoddyKit 上的免费 MLOps Academy 课时。这是第 3 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 MLOps Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 MLOps Academy 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

From Numbers to Pictures

Prometheus stores your metrics, but raw numbers are hard to read. Grafana turns them into charts you can actually understand at a glance. 📊

Grafana Plus Prometheus

Grafana does not store data itself. It queries a data source like Prometheus and draws whatever that source returns.

Add the Data Source

First step in Grafana is connecting Prometheus as a data source. You give it the Prometheus server URL and test the connection.

# Grafana > Connections > Data sources
# URL: http://prometheus:9090

Dashboards and Panels

A dashboard is a screen of panels. Each panel is one chart driven by one query, so you compose a full view from small pieces.

Query with PromQL

Panels ask Prometheus questions in PromQL, its query language. This is where you turn stored metrics into a meaningful line.

Chart the Request Rate

The rate function shows per-second change of a counter over a window. It is the classic way to plot traffic to your model.

rate(predictions_total[5m])

Chart the Error Ratio

Divide error rate by total rate to plot an error ratio. A rising line here means more requests are failing.

rate(errors_total[5m]) / rate(predictions_total[5m])

Chart p99 Latency

From your latency histogram, compute the 99th percentile so you see the slow tail, not just the average.

histogram_quantile(0.99, rate(predict_seconds_bucket[5m]))

Use Big Stat Panels

For single numbers like current error rate, a stat panel shows one large value and can flash red past a threshold. ⚠️

Variables Make It Reusable

Add a dashboard variable for model version, then one dashboard works for every model instead of copying it per version.

rate(predictions_total{model="$model"}[5m])

Save It as Code

Every dashboard can be exported to JSON. Commit that file to Git so your dashboards are versioned and reproducible like the rest of your stack.

Quick Check

You want a Grafana panel showing per-second request traffic to your model from a counter. Which PromQL fits?

Recap

You connected Grafana to Prometheus, built panels with PromQL for rate, errors, and p99 latency, and saved the dashboard as JSON in Git. ✅

常见问题解答

「构建 Grafana 仪表板」课时是免费的吗？

是的 — 「构建 Grafana 仪表板」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 MLOps Academy 课程的其余内容，请升级到 CoddyKit PRO。 MLOps Academy 课程共包含 4 节课。

「构建 Grafana 仪表板」这节课中我会学到什么？

一眼查看服务健康状况你通过在浏览器中直接运行的动手代码来练习 MLOps Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 MLOps Academy 需要有经验吗？

无需任何先前经验。CoddyKit 上的 MLOps Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 3 节课，共 4 节。

「构建 Grafana 仪表板」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 MLOps Academy 课中编写并运行代码吗？

能。每节 MLOps Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 MLOps Academy