OSI and TCP/IP Models

Why Models Matter

Networking models are not cables, switches, or commands. They are organized ways to describe what must happen for communication to work. The two models a CCST Networking candidate should recognize are the OSI model and the TCP/IP model. Both are useful because they separate a complicated event, such as loading a secure web page, into smaller functions that can be tested and discussed.

The OSI model has seven layers. Layer 1, Physical, deals with signals, connectors, cabling, radio, light, voltage, pinouts, and link presence. If an Ethernet cable is unplugged, a fiber strand is broken, or a wireless signal is too weak, start here. Layer 2, Data Link, handles local network delivery. Ethernet and Wi-Fi use frames and MAC addresses at this layer. Switch forwarding, wireless association, and many VLAN concepts fit here. Layer 3, Network, handles logical addressing and routing. IPv4, IPv6, subnet masks, prefixes, and default gateways are Layer 3 topics.

Layer 4, Transport, provides process-to-process communication using TCP or UDP ports. TCP adds connection setup, sequencing, acknowledgments, and retransmission; UDP is simpler and has less overhead.

Layers 5 through 7 are often discussed together in entry-level networking. Layer 5, Session, relates to managing conversations between systems. Layer 6, Presentation, relates to data representation, encryption, and formatting. Layer 7, Application, is where user-facing network services and application protocols operate, such as DNS, DHCP, HTTP, HTTPS, SMTP, IMAP, SSH, and SNMP. In day-to-day support, a technician may not label every upper-layer function separately, but the OSI names help when reading documentation or escalating to an engineer.

The TCP/IP model is closer to how Internet protocol suites are commonly grouped. A four-layer version includes network access, internet, transport, and application. Network access combines physical and data link functions such as Ethernet and Wi-Fi. The internet layer maps to IP addressing and routing. The transport layer maps to TCP and UDP. The application layer includes protocols that support user or system tasks, such as web access, name resolution, address assignment, email, file transfer, and management.

Encapsulation is the process of adding information as data moves down the layers before transmission. An application creates data. TCP may add a transport header and produce a segment. IP adds a network header and produces a packet. Ethernet or Wi-Fi adds data link information and produces a frame. The physical layer sends bits as electrical, optical, or radio signals. On the receiving device, decapsulation reverses the process so the receiving application gets usable data.

Layered troubleshooting keeps work disciplined. If a desktop has no link light, testing DNS first wastes time because the physical layer is already suspect. If link is good and the device has a valid IP address but cannot leave the subnet, the default gateway or routing path becomes more interesting. If ping to an IP address works but a website name fails, DNS is likely involved. If DNS resolves but only one application fails, the issue may be port filtering, server availability, credentials, certificate trust, or the application itself.

For CCST-level work, memorize the purpose of each OSI layer, understand the simpler TCP/IP grouping, and practice mapping symptoms to layers. The point is not reciting a model in isolation. The point is using the model to explain where traffic is failing and what evidence supports your next step.

Study Checkpoint

• Topic: OSI and TCP/IP Models.
• Verify the official Cisco concept before memorizing a shortcut.
• Practice the technician action: observe, document, test, fix when supported, or escalate.