GUIDEServer Management

Monitoring Server Performance (CPU, RAM, Disk, Network)

6 min read

Keeping a close eye on server performance is essential for maintaining a healthy, responsive infrastructure. Data Mammoth provides built-in monitoring tools in the dashboard alongside powerful command-line utilities that give you deep visibility into how your server resources are being consumed.

This guide covers how to read performance metrics in the dashboard and how to investigate resource usage from the command line.

Dashboard Performance Metrics

Your server dashboard includes real-time and historical graphs for the four primary resource categories. To access them:

Navigate to Servers and click on your server.

Scroll down to the Metrics or Performance section.

CPU Usage

The CPU graph shows the percentage of available processing power currently in use.

Normal range: 5% to 40% for most web applications during regular traffic.
Concerning range: Sustained usage above 80% may indicate the server needs optimization or more resources.
Critical range: Persistent 100% usage will cause slow response times, failed requests, and degraded service.

Common causes of high CPU usage include unoptimized database queries, runaway scripts, brute-force login attempts, or insufficient resources for your workload.

RAM (Memory) Usage

The memory graph shows how much of your allocated RAM is actively in use.

Important note for Linux servers: Linux aggressively caches disk data in unused RAM. This means the reported memory usage may appear high even when the server is not under load. What matters is the amount of memory available to applications, not the raw usage number.

To see true memory availability from the command line:

free -h

Look at the available column, which shows how much memory is truly available for new processes.

Disk Usage

The disk usage indicator shows the percentage of your storage that is consumed. Key thresholds to watch:

Below 70% — Healthy. No action required.
70% to 85% — Monitor actively. Plan for cleanup or expansion.
85% to 95% — Warning. Clean up unnecessary files, logs, or old backups.
Above 95% — Critical. The server may become unstable, fail to write logs, or crash.

Network Traffic

The network graph displays bandwidth usage over time, split into inbound (download) and outbound (upload) traffic.

Inbound traffic — Data arriving at your server, such as incoming web requests, file uploads, and API calls.
Outbound traffic — Data leaving your server, such as web page responses, file downloads, and backup transfers.

Unexpected spikes may indicate a traffic surge, a DDoS attack, or an unauthorized process transferring data.

Command-Line Monitoring Tools

For deeper analysis, connect to your server via SSH and use these built-in tools.

htop — Interactive Process Monitor

htop

htop provides a real-time, interactive view of all running processes sorted by resource usage. It shows:

Per-core CPU usage bars
Memory and swap usage bars
Process list with PID, user, CPU%, MEM%, and command
Ability to sort, filter, and kill processes interactively

If htop is not installed, install it:

# Ubuntu/Debian
sudo apt install htop -y
CentOS/AlmaLinux
sudo dnf install htop -y

vmstat — Virtual Memory Statistics

vmstat 5 10

This runs vmstat every 5 seconds for 10 iterations, showing CPU, memory, I/O, and system activity. Key columns to watch:

us — User CPU time (application workload)
sy — System CPU time (kernel workload)
wa — I/O wait (time spent waiting for disk operations)
free — Available memory in KB

High wa values indicate disk I/O bottlenecks.

iostat — Disk I/O Statistics

iostat -x 5

Shows detailed disk I/O performance, including read/write speeds, I/O operations per second (IOPS), and utilization percentage. Useful for diagnosing slow disk performance.

df — Disk Space Usage

df -h

Displays disk space usage for all mounted filesystems in human-readable format. Check the Use% column to identify partitions that are running low on space.

du — Directory Size

du -sh /var/log/*
du -sh /home/*

Shows how much space individual directories are consuming. Useful for finding large files or directories that are filling up your disk.

nethogs — Per-Process Network Usage

sudo nethogs

Shows network bandwidth usage broken down by process. Helpful for identifying which application is consuming the most bandwidth.

Install it with:

# Ubuntu/Debian
sudo apt install nethogs -y
CentOS/AlmaLinux
sudo dnf install nethogs -y

Setting Up Monitoring Alerts

While the Data Mammoth dashboard provides visual monitoring, you may want to receive proactive notifications when resource thresholds are crossed. Consider these approaches:

Dashboard alerts — Check your dashboard settings for built-in alert options that notify you via email when CPU, RAM, or disk usage exceeds a threshold.

Cron-based scripts — Write a simple bash script that checks resource usage and sends an email alert when thresholds are crossed.

Third-party monitoring — Tools like Uptime Robot, Netdata, or Zabbix can be installed on your server for advanced monitoring and alerting.

Common Performance Issues and Solutions

High CPU Usage

Run htop and sort by CPU to identify the top consumer.

Check for runaway processes or infinite loops.

Optimize database queries if MySQL/PostgreSQL is the top consumer.

Consider upgrading your plan if the workload legitimately needs more CPU. See How to Upgrade Your Server Plan.

High Memory Usage

Run free -h to check actual available memory.

Use htop sorted by memory to identify top consumers.

Check for memory leaks in long-running applications.

Tune application settings (e.g., reduce Apache/Nginx worker processes or PHP memory limits).

Disk Space Running Out

Run df -h to identify the full partition.

Check log files: du -sh /var/log/* — rotate or clear old logs.

Remove old backups or temporary files.

Clear package manager cache: sudo apt clean or sudo dnf clean all.

High Network Usage

Run sudo nethogs to find the process using bandwidth.

Check for unauthorized processes or malware.

Review backup schedules — large backups can consume significant bandwidth.

Consider enabling DDoS protection if you suspect an attack. See DDoS Protection.

What to Do Next

Understanding Your Server Dashboard — Navigate all dashboard features.
How to Upgrade Your Server Plan — Scale up when you need more resources.
Server Firewall Hardening Guide — Secure your server against unauthorized access.
Complete Server Security Checklist — Review your full security posture.

Was this article helpful?

← Back to Server Management Browse all categories →

Still have questions?

Contact Support →Submit a Ticket