Networking Protocols Quiz

Chapter 5: Link Layer

• This chapter discusses the link layer, which is responsible for transferring datagrams between physically adjacent nodes on a network.
• The link layer uses various protocols to manage communication channels and provide services such as error detection and correction.
• Terminology: Hosts and routers are referred to as nodes. Communication channels are called links. Links can be wired or wireless. A layer-2 packet is a frame, encapsulating a datagram.
• Link Layer Services:
  • Framing and Link Access: Encapsulates datagrams into frames by adding headers and trailers. Used to access shared mediums ("MAC" addresses identify source and destination). Different than IP addressing.
  • Reliable Delivery: This is less commonly used for low bit-error links. Wireless links use both link-level and end-end reliability.
  • Flow Control: Manages the pacing between sending and receiving nodes on a link.
  • Error Detection: Detects errors caused by signal attenuation or noise. Methods like parity checks and cyclic redundancy checks (CRCs) are used to find and correct errors.
  • Error Correction: Identifies and corrects errors without requiring retransmission.
  • Half-duplex and Full-duplex: Half-duplex allows transmission in one direction at a time, while full-duplex allows transmission in both directions simultaneously.
• Location: The link layer is implemented in a network interface card (NIC) or on a chip within each host.
• Adaptors (NICs) communicate by encapsulating datagrams in a frame, adding error detection bits (or other error correction), and performing rdt (reliable data transfer) and flow control.
• Link Layer, LANs Outline: Covers topics like introduction, services, error detection, correction; multiple access protocols (e.g. TDMA, FDMA); and local area networks (LANs) including addressing, ARP, Ethernet, switches, and Virtual LANs (VLANs). Also includes link virtualization (MPLS), data center networking, and a case study on a typical web request.
• MAC Addresses and ARP: Each network interface (NIC) has a unique 48-bit MAC address. ARP (Address Resolution Protocol) is used to translate an IP address to a MAC address, and vice-versa, allowing communication within a local area network.
• Ethernet: A widely used wired LAN technology. It uses a shared bus or star topology, CSMA/CD (Carrier Sense Multiple Access/Collision Detection) for access protocols;
  • Ethernet's physical topology varies from bus to star.
  • An Ethernet frame structure has a preamble, destination and source addresses, type, data, and CRC (Cyclic Redundancy Check) fields.
• Ethernet, Reliable and Connectionless. Ethernet is unreliable and connectionless, meaning it doesn't use handshaking between sending and receiving NICs, and receiving NICs don't send acks or nacks to sending NICs.
  • For data in dropped frames, the higher layer protocol like TCP adds retransmission methods to ensure reliability.
• 802.3 Ethernet Standards: Details many different Ethernet standards, speeds (2 Mbps to 10 Gbps), and physical layers like fiber and cable.
• Ethernet Switch: An active link layer device that stores and forwards frames. It examines MAC addresses in incoming frames to selectively forward them to the appropriate outgoing links. A switch uses CSMA/CD to access the segment. Switches are transparent, meaning hosts are unaware of their presence, and plug-and-play, meaning they can be used without needing configuration. Switches also use self-learning to build their forwarding tables.
• Switch Forwarding Tables: A switch maintains a forwarding table mapping MAC addresses to ports. This table is dynamically updated as the switch learns which hosts are reachable through which interfaces in the LAN.
• Switch Self-Learning: A switch's ability to learn the MAC addresses of hosts on its connected network segments and create a forwarding table to efficiently forward frames.
• Interconnecting Switches: Switches can be interconnected together to create larger networks.
• Data Center Networks: Details the architecture of networks in data centers and the challenges in design to support the ever-increasing number of high-speed connections and traffic. Load balancing and redundancy are key concepts that are discussed. A section on the typical interactions involved in a web server request is reviewed to exemplify the protocols mentioned in this chapter.
• MPLS (Multiprotocol Label Switching): A high-speed IP forwarding technique that uses fixed-length labels instead of IP addresses for fast lookup and forwarding.
• MPLS Signaling: Describes the protocols and techniques involved in network configuration for forwarding to ensure data can transition across different LANs.
• VLANs (Virtual Local Area Networks): A networking technology that allows the creation of multiple logical networks (VLANs) over a single physical network infrastructure. Using VLANs improves network security and efficiency.

Additional Topics (from Chapter 5, as identified in the discussion)

• Wireless, Multimedia, Network Security, Network Management. These topics are not further explored in the current chapter but may be discussed in later chapters.