Network fundamentals are the invisible infrastructure that make every web application, API call, and cloud service work. For a developer studying for the 200-901 exam, understanding how data travels across a network is essential for debugging connectivity issues, writing efficient code, and designing resilient systems. Without this knowledge, your application is like a letter with no address—it will never reach its destination.
Jump to a section
A simple way to picture Network Fundamentals for Developers
An airport baggage handling system is a precise, layered model of communication between completely separate systems. Suitcases are the data. The conveyor belts are the physical medium. The barcode tags are the network-layer addresses. The sorting algorithms are the routing protocols. The baggage claim carousel is the final delivery point.
When you check a bag at the counter in London Heathrow, the airline's system wraps your suitcase in a physical tag. That tag contains a destination code (like an IP address). The conveyor belt (Layer 1) physically moves the bag. The barcode scanners read the tag (Layer 2) to ensure the bag goes to the correct sorting area. The airport's central computer decides which flight the bag belongs on (Layer 3 routing). The baggage handlers at the gate verify the bag's tag against the passenger manifest (Layer 4 session). The bag is loaded onto the plane (Layer 5 session continuation). The flight crew ensures the bag is secure (Layer 6 presentation for encryption). The passenger finally claims the bag at the carousel (Layer 7 application).
If the tag is damaged or the sorting system misreads the code, the bag never arrives. Similarly, if a network packet loses its frame or routing information, data never reaches the destination. The baggage system works because each layer has a specific job and trusts the layers below it to do their job correctly. This is exactly how the OSI model operates for network communication.
The OSI (Open Systems Interconnection) model is a conceptual framework that divides network communication into seven distinct layers. Each layer has a specific function and talks only to the layer above and below it. Think of it as the rulebook for how data moves from one computer to another. The seven layers are: Physical, Data Link, Network, Transport, Session, Presentation, and Application.
Layer 1, the Physical layer, deals with the actual hardware—cables, switches (hardware that connects devices on the same network), and radio waves. It sends raw bits as electrical signals or light pulses. Layer 2, the Data Link layer, organises these bits into frames (structured packets of data with a header and trailer) and adds hardware-level addressing (MAC addresses, which are unique identifiers burned into network interface cards). Switches operate at this layer, forwarding frames within a local network.
Layer 3, the Network layer, is where logical addressing (IP addresses—unique identifiers assigned to devices on a network) comes in. Routers (devices that forward data between different networks) work here. This layer decides the best path for data to travel from source to destination using routing protocols like OSPF (Open Shortest Path First). Layer 4, the Transport layer, ensures reliable delivery through protocols like TCP (Transmission Control Protocol, which guarantees data arrives intact) and UDP (User Datagram Protocol, which is faster but doesn't guarantee delivery). This layer manages segmentation (breaking data into smaller chunks) and reassembly (putting them back together).
Layers 5, 6, and 7 are considered the 'upper layers'. Layer 5, the Session layer, manages the dialogue between applications—establishing, maintaining, and terminating connections. Layer 6, the Presentation layer, handles translation (converting data between formats), encryption (scrambling data for security), and compression (reducing data size). Layer 7, the Application layer, is what end-user software interacts with—HTTP for web browsing, SMTP for email, and DNS (Domain Name System) for converting domain names to IP addresses.
Network devices also have distinct roles. A hub (obsolete) blindly repeats signals. A switch learns which MAC addresses are on each port and only forwards frames to the correct destination. A router examines the IP address and decides the best path. A firewall (security device) filters traffic based on rules. A load balancer distributes incoming traffic across multiple servers.
In practice, when you type a URL into a browser, the following happens: DNS resolves the domain to an IP address. TCP establishes a reliable connection. HTTP sends a request as formatted text. The network layer packets travel across routers. The data link frames hop across switches. The physical layer cable carries the electrical signal. Finally, the web server's application layer interprets the request and sends back the response through the same layers in reverse.
For the 200-901 exam, you must understand which devices operate at which layers, the naming conventions for data at each layer (bits, frames, packets, segments), and the purpose of each layer. You should also know common protocols: HTTP, HTTPS, TCP, UDP, IP, DNS, DHCP (Dynamic Host Configuration Protocol, which assigns IP addresses automatically), and ARP (Address Resolution Protocol, which maps IP addresses to MAC addresses).
The OSI model is not how the internet actually works today—the internet uses the TCP/IP model, which combines layers (Application, Transport, Internet, Network Access). But the OSI model is still taught because it helps developers reason about network problems at the correct abstraction level. For example, if you cannot reach a website, you can isolate whether the issue is at the physical layer (cable unplugged), the network layer (wrong IP address), or the application layer (web server down).
Physical Connection
The process begins when your computer's network interface card (NIC) sends electrical signals or light pulses through a cable (or radio waves for Wi-Fi). This is the Physical layer (Layer 1). Without a functioning physical connection, no data can be transmitted. A faulty cable, disconnected plug, or dead Wi-Fi antenna stops everything.
Data Link Framing
The Data Link layer (Layer 2) organises the raw bits into frames. It adds the source and destination MAC addresses. Your computer's NIC has a unique MAC address burned into it. The switch reads this destination MAC and forwards the frame only to the correct device on the local network. If the MAC address is unknown, the switch broadcasts the frame to all devices.
Network Layer Routing
Once the frame reaches the router, the router strips off the frame header to reveal the IP packet. The router examines the destination IP address and consults its routing table (a list of known paths) to decide the best next hop. This is the Network layer (Layer 3). The router then re-encapsulates the packet into a new frame and sends it onward.
Transport Layer Segmentation
At the Transport layer (Layer 4), TCP divides the data into segments and assigns sequence numbers. This ensures that even if segments arrive out of order, the receiver can reassemble them correctly. TCP also establishes a three-way handshake (SYN, SYN-ACK, ACK) to open a reliable connection before any data flows. UDP skips this handshake and simply sends datagrams.
Upper Layers (Session, Presentation, Application)
At the destination device, the Session layer (Layer 5) manages the ongoing dialogue between applications. The Presentation layer (Layer 6) decrypts, decompresses, or converts the data if needed. Finally, the Application layer (Layer 7) makes the data available to the end-user software, such as a web browser rendering an HTTP response or an email client displaying a message.
Response and Reverse Path
The destination server sends a response back through the same layers in reverse order: from Application down to Physical. Each layer adds its own header or performs its function. This bidirectional flow is what enables interactive applications like web browsing, video calls, and file transfers to work seamlessly.
An IT professional troubleshooting a slow web application in a corporate network uses network fundamentals every day. Imagine you are a developer supporting a cloud-based inventory management system hosted on Amazon EC2 (virtual servers). Users report that the application is intermittent—sometimes it loads instantly, other times it times out.
First, you check the physical layer. Is the user's Wi-Fi router working? Are there any loose ethernet cables in the server room? This is rare, but rule it out. Next, you examine the data link layer using commands like 'arp -a' (to view the MAC address table). You discover that the user's laptop is connected to a switch port that is flapping between two VLANs, causing constant MAC address relearning. You fix this by configuring the switch port to a single VLAN.
Then you move to the network layer. You ping the EC2 instance's public IP address. If it fails, you know the problem is beyond the local network. You then use 'tracert' (traceroute) to see which router hop is dropping packets. You find that one intermediate router has 30% packet loss. This is a routing issue that the cloud provider needs to fix. If the ping succeeds but users still see slowness, you check the transport layer (TCP) for retransmission patterns using tools like Wireshark (a network packet analyser). High retransmission indicates packet loss due to network congestion or a misconfigured load balancer.
You then review the application layer logs. The web server logs show HTTP 503 errors, meaning the server is overwhelmed. You adjust the load balancer's health check interval and scaling policies. Finally, you check DNS—a misconfigured DNS record could be sending traffic to the wrong server entirely.
In this scenario, the IT professional uses knowledge of OSI layers to systematically isolate the fault. They also use practical tools like ping (Layer 3), traceroute (Layer 3), ARP (Layer 2), Wireshark (Layers 2-7), and netstat (Layer 4). Without understanding network fundamentals, you would be randomly clicking settings and praying for a fix. Instead, you follow the layered model and resolve the issue in minutes.
Common tools used in real-world networking include: - 'ping': tests connectivity at Layer 3 - 'traceroute': maps the path packets take across routers - 'netstat': displays active connections and their state at Layer 4 - 'ipconfig' (Windows) or 'ifconfig' (Linux): shows IP address, subnet mask, and default gateway - 'nslookup' or 'dig': queries DNS servers for domain resolution - 'Wireshark': captures and analyses packets at all layers - 'telnet' or 'nc' (netcat): tests connectivity to a specific port (Layer 4)
The 200-901 exam tests 6.1 'Describe the purpose and function of network devices, protocols, and the OSI model' through multiple-choice questions that focus on matching, ordering, and scenario-based recall. You will NOT be asked to memorise every OSI layer in order from memory—but you must recognise which layer corresponds to which function. The exam loves to test your ability to identify which device belongs to which layer and which protocol operates at which level.
Specific concepts that appear frequently:
Order of OSI layers: Physical, Data Link, Network, Transport, Session, Presentation, Application. A trap is listing 'Network' before 'Data Link'.
Function mapping: Which layer does routing? (Layer 3). Which layer handles flow control? (Layer 4). Which layer converts data formats? (Layer 6).
Device mapping: Routers work at Layer 3, switches at Layer 2, hubs at Layer 1, firewalls can work at Layers 3 and 4, load balancers at Layer 4 or 7.
Data unit names: Bits at Layer 1, frames at Layer 2, packets at Layer 3, segments at Layer 4.
Protocol mapping: HTTP (Layer 7), HTTPS (Layer 7 with encryption at Layer 6), TCP (Layer 4), UDP (Layer 4), IP (Layer 3), DNS (Layer 7 but uses Layer 4), DHCP (Layer 7 but uses Layer 2/3).
Comparing TCP vs UDP: TCP is connection-oriented, reliable, ordered; UDP is connectionless, unreliable, unordered. The exam may ask which protocol to use for streaming video (UDP, because speed matters more than retransmission) versus file transfer (TCP).
The exam also includes traps such as: - 'The Presentation layer is responsible for encryption' is correct, but some questions say 'Session layer handles encryption' which is false. - 'A switch operates at Layer 2' is true, but if the switch performs routing functions (multilayer switch), it also operates at Layer 3. The exam may test whether you know that a standard switch is Layer 2 only. - 'ARP belongs to the Network layer' is a common trick—ARP actually belongs to the Data Link layer (it maps IP to MAC). - 'The Physical layer sends data as packets' is false; it sends raw bits.
To memorise effectively: create a mnemonic for the OSI layers (e.g., 'Please Do Not Throw Sausage Pizza Away'). Practise associating each layer with its core function and one example device. Also understand that the TCP/IP model (Application, Transport, Internet, Network Access) is what the internet actually uses, but the exam expects you to know the OSI model explicitly as the conceptual framework.
Key definitions to memorise:
MAC address: a 48-bit hardware address burned into a network interface card (NIC)
IP address: a 32-bit (IPv4) or 128-bit (IPv6) logical address assigned to a device
Subnet mask: determines which part of an IP address is the network portion and which is the host portion
Default gateway: the router that connects a local network to other networks
Firewall: a device that filters traffic based on a set of security rules
Load balancer: distributes incoming traffic across multiple servers to optimise performance and prevent overload
The OSI model has seven layers: Physical, Data Link, Network, Transport, Session, Presentation, Application, each with a specific function in network communication.
Routers operate at Layer 3 (Network) using IP addresses to forward packets between different networks, while switches work at Layer 2 (Data Link) within a single network using MAC addresses.
TCP guarantees reliable data delivery with acknowledgements and retransmission; UDP is faster but does not guarantee delivery, making it ideal for real-time applications like video streaming.
Data at each layer has a specific name: bits at Layer 1, frames at Layer 2, packets at Layer 3, and segments at Layer 4.
Common protocols are assigned to specific layers: HTTP and DNS are Layer 7; TCP and UDP are Layer 4; IP is Layer 3.
The internet uses the TCP/IP model with four layers (Application, Transport, Internet, Network Access), but the OSI model is essential for troubleshooting and exam questions.
Network devices include hubs (Layer 1), switches (Layer 2), routers (Layer 3), firewalls (Layers 3-4), and load balancers (Layers 4 or 7).
These come up on the exam all the time. Here's how to tell them apart.
TCP (Transmission Control Protocol)
Connection-oriented: requires a three-way handshake before data transfer
Guarantees delivery through acknowledgements and retransmission of lost packets
Used for applications where reliability matters: web browsing (HTTP), email (SMTP), file transfer (FTP)
UDP (User Datagram Protocol)
Connectionless: no handshake required, data is sent immediately
No guarantee of delivery; packets may be lost or arrive out of order
Used for applications where speed matters and loss is tolerable: video streaming, VoIP, online gaming
Router
Operates at Layer 3 (Network) using IP addresses
Connects different networks (e.g., LAN to the internet)
Maintains a routing table to determine the best path for packets
Switch
Operates at Layer 2 (Data Link) using MAC addresses
Connects devices within the same local network (LAN)
Learns and stores MAC addresses in a forwarding table to direct frames to specific ports
IPv4
32-bit address space, allowing approximately 4.3 billion unique addresses
Written in dotted decimal format (e.g., 192.168.1.1)
Uses ARP (Address Resolution Protocol) to map IP addresses to MAC addresses
IPv6
128-bit address space, providing a virtually unlimited number of addresses
Written in hexadecimal colon format (e.g., 2001:0db8::1)
Uses Neighbor Discovery Protocol instead of ARP
Firewall
Filters traffic based on defined security rules (allow/deny)
Operates at Layer 3 and Layer 4 (IP and port-based filtering)
Protects a network from unauthorised access and attacks
Load Balancer
Distributes incoming traffic across multiple servers to avoid overload
Operates at Layer 4 (TCP/UDP) or Layer 7 (HTTP/HTTPS)
Improves application availability and performance
MAC Address
48-bit hardware address burned into a device's NIC by the manufacturer
Used for communication within a local network (Data Link layer)
Typically remains unchanged for the life of the device
IP Address
32-bit (IPv4) or 128-bit (IPv6) logical address assigned by the network
Used for routing packets across different networks (Internet layer)
Can change based on network configuration (DHCP, static assignment)
Mistake
The OSI model is how the internet actually works, and all data must pass through all seven layers in order on every communication.
Correct
The internet uses the TCP/IP model, which has only four layers. The OSI model is a theoretical framework for teaching and troubleshooting. Data may skip layers in practice—for example, a direct cable connection bypasses Network and Transport layers.
Beginners see seven layers in textbooks and assume they are all strictly followed. The TCP/IP model is simpler and more accurate, but the OSI model is still taught because it helps isolate problems.
Mistake
A router and a switch are basically the same thing, just different brand names.
Correct
A router operates at Layer 3 (Network) and connects different networks using IP addresses. A switch operates at Layer 2 (Data Link) and connects devices within the same network using MAC addresses. They perform fundamentally different jobs.
Both devices look similar (metal boxes with blinking lights) and are often marketed together, leading beginners to think they are interchangeable.
Mistake
TCP and UDP are both protocols that guarantee data delivery, just at different speeds.
Correct
Only TCP (Transmission Control Protocol) guarantees delivery through acknowledgements and retransmission. UDP (User Datagram Protocol) is 'fire and forget'—it sends data without any delivery confirmation. Video streaming can tolerate some packet loss, so UDP is preferred for speed.
The word 'protocol' sounds technical, and many beginners assume all protocols provide the same level of reliability. The difference is critical for application design.
Mistake
A MAC address is the same as an IP address, just used for different purposes interchangeably.
Correct
A MAC address is a permanent hardware identifier assigned by the manufacturer, used for local network communication. An IP address is a logical address assigned by the network, used for routing across the internet. They serve different layers (Layer 2 vs Layer 3) and cannot be swapped.
Both are 'addresses', so beginners conflate them. The fact that devices often have both simultaneously adds to the confusion.
Mistake
The Presentation layer's only job is to encrypt data.
Correct
The Presentation layer (Layer 6) handles translation (converting data between formats like JPEG to BMP), encryption (like SSL/TLS for secure web traffic), and compression (reducing file size before transmission). It is not just encryption.
Encryption is a prominent security topic, so beginners latch onto that function and forget the other responsibilities of the layer.
Mistake
DNS is a protocol that works at the Network layer because it resolves IP addresses.
Correct
DNS (Domain Name System) is an Application layer (Layer 7) protocol. It uses TCP or UDP at the Transport layer, but the service itself (querying a DNS server) is an application function that translates human-friendly names to IP addresses.
Because DNS maps names to IP addresses, and IP addresses are Layer 3, beginners incorrectly place DNS at the Network layer.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
The OSI model is a seven-layer framework that describes how data travels through a network. For the 200-901 exam, you need to know the order of layers, the function of each, and which devices and protocols operate at each layer to answer scenario-based questions.
TCP is connection-oriented and guarantees reliable delivery through acknowledgements and retransmission. UDP is connectionless and does not guarantee delivery, making it faster but less reliable. Use TCP for file transfers and UDP for streaming video.
A MAC address is a permanent hardware identifier (like a serial number) burned into a network interface card, used for local network communication (Layer 2). An IP address is a logical address assigned by the network, used for routing across the internet (Layer 3).
A router operates at Layer 3 (the Network layer). It uses IP addresses to decide the best path for forwarding packets between different networks.
A switch is intelligent—it learns which MAC address is on each port and forwards frames only to the correct destination (Layer 2). A hub blindly repeats every signal to all ports (Layer 1), which wastes bandwidth and causes collisions.
Yes, DNS (Domain Name System) operates at Layer 7 (Application layer). It translates human-readable domain names (like google.com) into machine-readable IP addresses. It uses UDP or TCP at the Transport layer for transmission.
You've finished Network Fundamentals for Developers. Continue through the 200-901 study guide to build a complete picture of the exam.
Done with this chapter?