Server
About 5 min read
Last updated: 2026-04-16
What Is a Server
A server is a computer or software that provides services and data to other computers (clients) over a network. The term comes from "to serve" - servers receive requests and return processed results.
When you enter a URL in your browser, a web server receives the request and sends back HTML, images, and other content. Email delivery, domain name resolution, database queries - virtually every internet service relies on servers behind the scenes.
The word "server" can refer to both hardware (a physical machine) and software (Apache, Nginx, etc.). A single physical machine commonly runs multiple server software instances simultaneously.
Types of Servers
In practice, a single server may serve multiple roles, or a single role may be distributed across hundreds of servers. Large-scale web services typically separate web, application, and database servers, scaling each independently.
Physical Servers vs Cloud
Serverless architectures (e.g., AWS Lambda) let developers build applications without managing servers. However, physical servers still run behind the scenes - "serverless" refers to the transfer of management responsibility, not the absence of servers.
Server Security Fundamentals
- Patch management: Apply security patches promptly. Unpatched known vulnerabilities are the biggest risk. Automate updates to minimize delay.
- Firewall configuration: Open only necessary ports and block access to unused services. Changing SSH from port 22 to a custom port reduces automated scanning.
- Access control: Disable root login, minimize sudo privileges, and follow the IAM principle of least privilege.
- Log monitoring: Regularly review access, authentication, and error logs. SIEM tools enable automated analysis.
- Backups: Follow the 3-2-1 rule. Offline backups are essential for ransomware protection.
Choosing the Right Server
- Traffic volume: A small site with a few thousand monthly page views can use shared hosting. Millions of monthly views require CDN integration and auto-scaling.
- Availability requirements: 99.9% (about 8.7 hours downtime/year) vs 99.99% (about 52 minutes/year) have vastly different redundancy costs.
- Data residency: GDPR and other regulations may require data to remain in specific regions, affecting cloud region selection.
- Operational capacity: Managed services and serverless reduce operational burden but limit customization.
"Default to cloud" isn't always optimal. For predictable, stable workloads, reserved instances or dedicated servers may be more cost-effective. Clarify requirements before comparing options.
Common Misconceptions
- Servers must be special high-performance computers
- Technically, a home PC or Raspberry Pi can function as a server. 'Server' describes a network role, not hardware specifications. Commercial environments use server-grade hardware for reliability and redundancy.
- Moving to the cloud eliminates server management
- With IaaS (e.g., EC2), OS patching and security configuration remain your responsibility. Even with PaaS and serverless, application-level security, access control, and monitoring are your job. Understanding the <a href='/glossary/shared-responsibility-model'>shared responsibility model</a> is critical.
- Serverless means no servers exist
- Serverless means the cloud provider manages the servers. Physical servers still run behind the scenes. The term describes an operational model where developers don't handle provisioning or scaling.