Cloud Monitoring with Prometheus and Grafana: Setup & Best Practices

Cloud Monitoring with Prometheus and Grafana Featured img BDCC

In today’s cloud-native era, infrastructure monitoring has become more than just a support function; it’s the core of performance, uptime, and reliability. Organizations are increasingly adopting open-source observability tools to monitor their systems efficiently. Among them, Prometheus and Grafana stand out as a powerful combination for cloud monitoring.

If you’re building scalable applications on the cloud, setting up real-time visibility into your infrastructure is non-negotiable. This blog takes a deep dive into how you can implement Cloud monitoring with Prometheus and Grafana, from setting them up to creating meaningful dashboards.

Why Monitoring Cloud Infrastructure Matters?

Modern applications run on distributed, dynamic, and containerized environments. From Kubernetes clusters to serverless functions, every component generates metrics that can help detect issues before they impact users. But to leverage those metrics, you need to:

Collect them efficiently
Store them with minimal overhead
Visualize them meaningfully
Alert on anomalies

That’s exactly where Prometheus and Grafana shine.

What is Prometheus?

Prometheus is an open-source systems monitoring and alerting toolkit originally built by SoundCloud. It’s designed to collect metrics from your services and store them in a time-series database.

Key Features

Pull-based metrics collection via HTTP endpoints
Multidimensional data model (labels, key-value pairs)
Built-in query language (PromQL)
Alerting via Alertmanager
Seamless integration with Kubernetes, Docker, and cloud services

What is Grafana?

Grafana is a powerful visualization and analytics tool that integrates with Prometheus and other data sources to present metrics on custom dashboards.

Grafana Highlights

Beautiful, interactive dashboards
Query editor with auto-complete
Flexible panel types: graphs, heatmaps, tables, gauges
Alerting based on thresholds or Prometheus queries
User access management and sharing options

When you combine Prometheus and Grafana, you get full-stack observability — collection, querying, visualization, and alerting — all in one seamless flow.

Prometheus + Grafana Architecture

Before diving into setup, let’s understand how they work together:

Prometheus scrapes metrics from cloud services (e.g., EC2, EKS, GKE) or exporters (e.g., node_exporter, blackbox_exporter).
These metrics are stored in Prometheus’ TSDB.
Grafana queries Prometheus to visualize metrics in real-time dashboards.
Alertmanager (optional) manages notifications based on defined rules.

Step-by-Step: Setting Up Prometheus for Cloud Monitoring

Let’s go through a simplified Prometheus setup for monitoring a cloud environment.

Step 1: Install Prometheus

You can deploy Prometheus on a VM or in a Kubernetes cluster. Here’s a quick command to get it running using Docker:

docker run -d -p 9090:9090 \
-v /path/to/prometheus.yml:/etc/prometheus/prometheus.yml \
prom/prometheus

Alternatively, in Kubernetes:

kubectl create namespace monitoring
helm install prometheus prometheus-community/prometheus -n monitoring

Step 2: Configure Targets in prometheus.yml

Here’s a simple example for monitoring EC2 instances:

scrape_configs:
- job_name: 'node'
static_configs:
- targets: ['your-ec2-ip:9100']

Make sure you install node_exporter on your instances to expose metrics.

Step 3: Verify Prometheus is Scraping Data

Access Prometheus at http://localhost:9090 and use PromQL to run queries like:

up
node_cpu_seconds_total
node_memory_MemAvailable_bytes

You’re now collecting infrastructure metrics — the first step to cloud observability.

Visualizing Data with Grafana Dashboards

Once Prometheus is feeding metrics, it’s time to create dashboards in Grafana.

Step 1: Install Grafana

You can deploy Grafana via Docker:

docker run -d -p 3000:3000 grafana/grafana

Or in Kubernetes:

helm install grafana grafana/grafana -n monitoring

Step 2: Add Prometheus as a Data Source

Go to Configuration → Data Sources
Select Prometheus
Enter the URL (e.g., http://localhost:9090) and save

Step 3: Import Pre-built Dashboards

Grafana supports importing community dashboards:

Go to + → Import
Use dashboard ID (e.g., 1860 for Node Exporter Full)
Choose your Prometheus data source

Step 4: Create Custom Grafana Panels

Use PromQL to build custom panels:

CPU usage:

100 - (avg by(instance) (irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)

Memory usage:

node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes * 100

These dashboards give you real-time visibility into your cloud infrastructure.

Use Cases: Cloud Monitoring with Prometheus and Grafana

Let’s explore practical use cases where this stack excels:

1. Kubernetes Monitoring

Native integration with kube-state-metrics and cAdvisor makes Prometheus ideal for Kubernetes observability. Grafana dashboards can track pod health, resource usage, and node status.

2. Multi-Cloud Infrastructure Monitoring

Whether you’re on AWS, Azure, or GCP — you can scrape metrics from EC2, RDS, GKE, AKS, etc., using exporters like:

cloudwatch_exporter (AWS)
azure_metrics_exporter (Azure)
gcp_exporter (GCP)

3. API and Application Monitoring

Use blackbox_exporter to test HTTP endpoints, SSL certs, DNS resolution, and latency metrics — essential for app uptime.

Alerting with Prometheus and Grafana

Monitoring without alerting is incomplete.

Prometheus Alertmanager:

Define rules in alert.rules.yml
Example:

- alert: InstanceDown
expr: up == 0
for: 1m
labels:
severity: critical
annotations:
summary: "Instance {{ $labels.instance }} down"

Integrate with Slack, PagerDuty, Email, etc.

Grafana Alerting (New):

Grafana 9+ supports unified alerting — create alerts right from dashboards and manage them centrally.

Best Practices for Cloud Observability

To make the most of your monitoring stack:

Use label-based metrics wisely to avoid cardinality explosion.
Apply recording rules to precompute expensive queries.
Leverage Grafana folders and permissions for team collaboration.
Export dashboards as JSON for version control.
Centralize logs with Loki or Fluentd for complete observability.

Prometheus Grafana Tutorial Summary

Feature	Prometheus	Grafana
Role	Metrics collection & alerting	Visualization & dashboarding
Integration	Kubernetes, EC2, custom exporters	Prometheus, InfluxDB, Loki, etc.
Query Language	PromQL	PromQL, SQL-like UI queries
Deployment	Docker, Kubernetes, Bare metal	Docker, Kubernetes, Cloud-hosted
Alerting	Yes (Alertmanager)	Yes (from v9+)

Final Thoughts: Why Use Prometheus + Grafana for Cloud Monitoring

Cloud monitoring with Prometheus and Grafana offers a flexible, cost-effective, and powerful solution for modern infrastructure. Whether you’re managing a hybrid cloud, microservices, or large-scale Kubernetes deployments; this combo gives you the clarity and control to ensure uptime and performance.

By integrating Prometheus for metric collection and Grafana for insightful dashboards, you unlock full visibility into your infrastructure health; backed by real-time alerts and beautiful visualizations.

Monitoring Cloud Infrastructure with Prometheus + Grafana