Outils de supervision côté serveur

When you run performance tests, it's not enough to just see if your application responds. You also need to know what's happening behind the scenes on your servers.

Server-side monitoring helps you understand if your servers are struggling under load, or if they have enough resources to handle user traffic.

Server-side monitoring involves collecting real-time data from your application servers and infrastructure. This data helps identify bottlenecks and performance issues.

• CPU Usage: Is the processor overworked?
• Memory Usage: Is the server running out of RAM?
• Disk I/O: Are read/write operations slowing things down?
• Network Traffic: Is there too much data moving in or out?

These metrics are crucial for accurate performance analysis.

Prometheus is a popular open-source monitoring system. It's designed to collect and store metrics as time series data, meaning data is stored with a timestamp.

Prometheus uses a pull model: it actively 'scrapes' or pulls metrics from configured targets (your servers or applications) at regular intervals.

In Prometheus, everything is a metric – a named set of time series data. Examples include http_requests_total or node_cpu_seconds_total.

To expose these metrics, applications or servers need exporters. An exporter is a small service that translates existing metrics from a system (like Linux servers or databases) into a format Prometheus can understand.

While Prometheus collects data, Grafana is the go-to tool for visualizing it. It's an open-source analytics and interactive visualization web application.

Grafana allows you to create dynamic dashboards with various panels (graphs, tables, heatmaps) to display your metrics beautifully and in real-time.

Grafana doesn't collect data itself; it connects to various data sources. Prometheus is one of its most common data sources.

Once connected, you can use Grafana's query editor to write PromQL (Prometheus Query Language) queries to select and manipulate your metrics, then display them on a dashboard.

New Relic is a comprehensive commercial observability platform. Unlike Prometheus, it's an all-in-one solution offering application performance monitoring (APM), infrastructure monitoring, log management, and more.

It provides deep insights into your entire software stack, from user experience to underlying infrastructure.

New Relic typically uses agents that you install directly on your servers or within your applications (e.g., Java agent, Python agent).

These agents automatically collect a vast amount of detailed performance data, sending it to the New Relic platform for analysis and visualization, often with minimal configuration.

Both Prometheus/Grafana and New Relic are powerful, but serve different needs:

• Prometheus + Grafana: Open-source, highly customizable, requires more setup/maintenance. Great for deep technical control and specific use cases.
• New Relic: Commercial, all-in-one, easier to set up for broad coverage, often includes advanced features and support. Better for quick, comprehensive insights across large systems.

Which of the following statements are true regarding server-side monitoring tools discussed?

You've explored the world of server-side monitoring! We covered:

• The importance of monitoring CPU, memory, disk, and network.
• Prometheus: An open-source, pull-based metric collection system with exporters.
• Grafana: An open-source tool for visualizing metrics from sources like Prometheus.
• New Relic: A commercial, agent-based, all-in-one observability platform.

Understanding these tools helps you diagnose server performance during load tests.