Podcast
Questions and Answers
The link layer is implemented solely in software without any hardware components.
The link layer is implemented solely in software without any hardware components.
False
Half-duplex communication allows both nodes to transmit simultaneously.
Half-duplex communication allows both nodes to transmit simultaneously.
False
Error correction can occur without retranmitting the original data.
Error correction can occur without retranmitting the original data.
True
Wireless links tend to have lower error rates compared to fiber optics.
Wireless links tend to have lower error rates compared to fiber optics.
Signup and view all the answers
Flow control is used to pace the transmission of data between adjacent nodes.
Flow control is used to pace the transmission of data between adjacent nodes.
Signup and view all the answers
The link layer is responsible for encapsulating the datagram into a segment before transmission.
The link layer is responsible for encapsulating the datagram into a segment before transmission.
Signup and view all the answers
MAC addresses are used in frame headers to identify the source and destination in the link layer.
MAC addresses are used in frame headers to identify the source and destination in the link layer.
Signup and view all the answers
The link layer only provides services over wired links and does not support wireless links.
The link layer only provides services over wired links and does not support wireless links.
Signup and view all the answers
The line between a local area network (LAN) and a wide area network (WAN) is defined solely by the geographical distance.
The line between a local area network (LAN) and a wide area network (WAN) is defined solely by the geographical distance.
Signup and view all the answers
Frames used in link layer transmissions can encapsulate multiple datagrams.
Frames used in link layer transmissions can encapsulate multiple datagrams.
Signup and view all the answers
In unslotted ALOHA, a node transmits immediately when a frame arrives.
In unslotted ALOHA, a node transmits immediately when a frame arrives.
Signup and view all the answers
The efficiency of slotted ALOHA is greater than that of pure ALOHA.
The efficiency of slotted ALOHA is greater than that of pure ALOHA.
Signup and view all the answers
If a Network Interface Card (NIC) detects a transmission while creating a frame, it continues with the transmission.
If a Network Interface Card (NIC) detects a transmission while creating a frame, it continues with the transmission.
Signup and view all the answers
CSMA/CD efficiency approaches 1 as the propagation delay ($T_{prop}$) goes to infinity.
CSMA/CD efficiency approaches 1 as the propagation delay ($T_{prop}$) goes to infinity.
Signup and view all the answers
CSMA requires nodes to defer their transmission if the channel is sensed busy.
CSMA requires nodes to defer their transmission if the channel is sensed busy.
Signup and view all the answers
Collisions in CSMA can never occur due to the listening protocol.
Collisions in CSMA can never occur due to the listening protocol.
Signup and view all the answers
Polling in 'taking turns' MAC protocols can lead to overhead and latency concerns.
Polling in 'taking turns' MAC protocols can lead to overhead and latency concerns.
Signup and view all the answers
In random access MAC protocols, a single node can utilize the channel fully during high load.
In random access MAC protocols, a single node can utilize the channel fully during high load.
Signup and view all the answers
In CSMA/CD, collision detection is challenging in wired LANs compared to wireless LANs.
In CSMA/CD, collision detection is challenging in wired LANs compared to wireless LANs.
Signup and view all the answers
In token passing protocols, a control token is passed sequentially from one node to the next.
In token passing protocols, a control token is passed sequentially from one node to the next.
Signup and view all the answers
Study Notes
Chapter 5: Link Layer
- This chapter covers the principles behind data link layer services, including error detection and correction, sharing a broadcast channel, and link layer addressing.
- It also discusses the instantiation and implementation of various link layer technologies, such as Ethernet, switched LANs, VLANs, and virtualized networks (MPLS).
- The concepts of MAC addresses and ARP are also introduced.
- The chapter concludes with a summary of MAC protocols and a detailed analysis of how a web request journeys down the protocol stack from application to link layer.
- A practical scenario of connecting a laptop to the internet is presented
Note on the Use of PPT Slides
- The slides are freely available for faculty, students, and readers.
- The slides are in PowerPoint format, including animations.
- Users can adapt the slides and content to their needs.
- If using the slides in a class, acknowledge their source.
- If publishing the slides on a website, acknowledge the source and copyright of the material.
Link Layer: Introduction
- Hosts and routers are nodes that connect adjacent nodes along the communication path, forming links.
- These links include wired links and wireless links, and encompass local area networks (LANs).
- A layer-2 packet is a frame, which encapsulates a datagram.
- The data link layer is responsible for transferring a datagram between physically adjacent nodes over a link.
Link Layer: Context
- Different link protocols transfer datagrams across various links.
- For example, Ethernet is used on the first link while frame relay is used on intermediate links, and 802.11 is used on the last link.
- Each link protocol offers distinct services, some of which may or may not provide reliable data transfer (rdt) over the link.
- A transportation analogy is used to illustrate how a trip functions like how datagrams transfer across links
Link Layer Services
- Framing and link access: encapsulate datagram into frame, adding header, trailer, and channel access on shared mediums.
- "MAC" addresses are used in frame headers to identify source and destination, different from IP addresses.
- Reliable delivery between adjacent nodes: this concept was introduced in a previous chapter.
- Wireless links have high error rates.
- Both link-level and end-to-end reliability are needed for wireless links.
Link Layer Services (More)
- Flow control: pacing between sending and receiving nodes.
- Error detection: errors may occur due to signal attenuation, noises. Receivers detect errors and signal sender for retransmission or frame drop.
- Error correction: receiver identifies and corrects bit errors without retransmission.
- Half-duplex and full-duplex: With half-duplex, nodes can transmit but not simultaneously, while in full-duplex, nodes can transmit simultaneously.
Where is the Link Layer Implemented?
- The link layer is implemented in each host's network adapter (NIC) or on a chip.
- Examples include Ethernet cards and 802.11 cards, as well as Ethernet chipsets.
- These components interface with the host's system buses.
- A combination of hardware, software, and firmware is involved in the implementation.
Adaptors Communicating
- Sending side encapsulates a datagram into a frame, adds error checking bits, implements reliable data transfer (rdt), and flow control.
MAC Protocols: Taxonomy
- MAC protocols are categorized into three classes:
- Channel partitioning: channels are divided into smaller pieces such as time slots, frequency bands, or codes. These pieces are then allocated to nodes for exclusive use.
- Random access: channels aren't divided, collisions are allowed, and nodes recover from these collisions.
- Taking turns: nodes take turns, and nodes with more transmissions can take longer turns.
Channel Partitioning MAC Protocols: TDMA
- TDMA (Time Division Multiple Access): access to the channel occurs in rounds, where each station gets a fixed-length slot in each round. Unused slots remain idle.
Channel Partitioning MAC Protocols: FDMA
- FDMA (Frequency Division Multiple Access): channels are divided into frequency bands, and each station is assigned a fixed frequency band. Unused time in these bands is considered idle.
Random Access Protocols
- Random access protocols are used when a node has a packet to send. Nodes transmit at full channel data rate without prior coordination, and collisions can occur.
- These protocols describe how to detect and recover from collisions, including slotted ALOHA, ALOHA, CSMA, CSMA/CD, and CSMA/CA.
Slotted ALOHA
- Assumptions: all frames are the same size; time is divided into equal-size slots (time to transmit one frame). Nodes start transmitting only at the beginning of a slot; nodes are synchronized. Collisions occur if two or more nodes transmit in the same slot.
- Operation: when a node gets a frame it transmits in the next slot. If there's no collision, it can send a new frame in the next slot; otherwise, the node retransmits the frame in each subsequent slot until it succeeds.
Slotted ALOHA Pros and Cons
- Pros: A single active node can continuously transmit at full channel rate, highly decentralized (slots in nodes need to be in sync), simple.
- Cons: Collisions can waste slots, and idle slots exist. Nodes may be able to detect collisions before transmission time of the packet. Clock synchronization is required for proper operation
Slotted ALOHA Efficiency
- Efficiency is the long-run fraction of successful slots with many nodes and frames.
- Theoretical maximum efficiency using this protocol is 1/e which is approximately 0.37.
Pure (Unslotted) ALOHA
- This is a simpler random access protocol with no synchronization. Frames arrive and are transmitted immediately.
- The collision probability increases as more frames are sent within a short interval.
Pure ALOHA Efficiency
- The efficiency of this protocol is even lower than slotted ALOHA at 1/(2e) or .18
CSMA (Carrier Sense Multiple Access)
- Nodes listen before transmitting. If the channel is idle, they transmit the entire frame; if busy, they defer transmission. It’s an analogy for not interrupting others.
CSMA Collisions
- Collisions can still occur because of propagation delay. Two nodes may not hear each other's transmissions.
- Collision results in entire packet transmission time being wasted.
CSMA/CD (Collision Detection)
- Collisions are detected within a short time, and colliding transmissions are aborted to reduce channel wastage.
- Collision detection is simple in wired LANs by measuring signal strength.
- It is more challenging in wireless LANs due to the strength of local transmissions overwhelming the received signal strength.
Ethernet CSMA/CD Algorithm
- NIC receives datagram from network layer, creates a frame. If the channel is idle, it starts transmitting the frame. If busy, it waits until the channel is idle.
- It transmits the entire frame without detecting any other transmissions.
- If a collision occurs, the NIC aborts the transmission, sends a jam signal, and enters a binary (exponential) back-off period where it calculates an appropriate delay interval before attempting to retransmit the frame.
CSMA/CD Efficiency
- The efficiency of the CSMA/CD protocol increases as propagation delay approaches zero, or transmission time approaches infinity.
“Taking Turns” MAC Protocols
- Channel partitioning MAC protocols divide and allocate channel resources to maintain efficiency at high load but become inefficient at low load.
- Random access protocols offer efficiency at low load but have high collision overhead at high load.
- "Taking turns protocols try to find a balance between these approaches.
Polling
- The master node "invites" slave nodes to transmit in turn. This approach is often used for dumb devices.
- Concerns include polling overhead, latency, and potential single points of failure (if the master fails).
Token Passing
- A control token is passed sequentially from one node to the next. This ensures a turn-based approach.
- Concerns include token overhead, latency, and single points of failure (if the token is lost).
Cable Access Network
- Internet frames, TV channels, and control information are transmitted downstream at different frequencies.
- Upstream Internet frames, TV control, and other information are transmitted upstream at different frequencies in time slots.
- DOCSIS (Data Over Cable Service Interface Specification) utilizes FDM (Frequency Division Multiplexing) and TDM (Time Division Multiplexing) over upstream and downstream channels.
Ethernet Switch
- Ethernet switches actively participate in frame handling.
- They store and forward Ethernet frames to appropriate segments based on destination MAC addresses.
- They use CSMA/CD to control access to the segment.
- They provide transparent functionality where hosts are unaware of their presence.
- Ethernet switches offer plug-and-play features and self-learning capabilities.
Switch: Multiple Simultaneous Transmissions
- Hosts have dedicated direct connections to switches.
- Ethernet protocols are used on each incoming link, preventing collisions.
- Each link is considered its own collision domain.
- Switching facilitates simultaneous transmissions without collisions.
Switch Forwarding Table
- Switches use forwarding tables to determine the appropriate interface for forwarding frames.
- Each entry in the table includes the MAC address of a host, the interface to reach the host, and a timestamp
Switch: Self-Learning
- Switches learn which hosts can be reached through which interfaces.
- When a frame is received, the switch learns the sender's location (LAN segment).
- The switch records the sender's location/segment pair in its switch table.
Switch: Frame Filtering/Forwarding
- Frames received by the switch are recorded, including the incoming link and MAC address of the sending host.
- The switch table is indexed using the destination MAC address.
- If the destination host is on the same segment as the arrival interface, the frame is dropped. Otherwise, the frame is forwarded to the interface indicated in the switch table.
- If no entry is in the table, the frame is flooded on all interfaces except the arriving one.
Self-Learning, Forwarding: Example
- When the destination is unknown, a frame is flooded.
- If the destination is known, the frame is selectively sent on the appropriate link.
Interconnecting Switches
- Switches can be interconnected.
- Self-learning capabilities are used in a similar way as in a single-switch configuration.
Data Center Networks
- Data centers contain tens or hundreds of thousands of closely coupled hosts.
- Common applications in data centers include e-business, content delivery, and search engines.
- Key challenges in data center networking include handling multiple applications, managing load balancing, and preventing bottlenecks.
Data Center Networks (Load Balancer)
- Load balancers manage and distribute application requests to internal clients, hiding data center internals from external clients.
- The main components include a load balancer, border router, access router, and tier 1, 2 switches along with server racks.
Data Center Networks (Interconnection)
- Switches, and server racks are interconnected extensively to increase throughput between racks, improve fault tolerance and efficiency.
Synthesis: A Day in the Life of a Web Request
- The journey down a protocol stack during a web request involves several steps of encapsulation, forwarding, and delivery.
- A scenario of a student requesting a web page from google.com is used to outline the steps and protocols involved in the request.
A Day in the Life: Scenario
- The scenario outlines the steps involved in a typical web request, starting from a user opening a browser and sending a request to a web server.
- The various components involved include the browser, school network, Comcast Network, DNS server and the google web server itself.
A Day in the Life: Connecting to the Internet
- DHCP is a crucial protocol for connecting to the internet. A laptop requests an IP address, router address, and DNS server address from the DHCP server via a broadcast frame to the router which encapsulates and delivers this response to the requesting client.
A Day in the Life: ARP (Before DNS, Before HTTP)
- This step occurs before DNS and HTTP. A query for the MAC address of the required resource is sent as a broadcast frame. The router will reply, allowing the client to send its request containing the DNS query, to the appropriate DNS server.
A Day in the Life: Using DNS
- The client's router sends a DNS query containing the desired web page's name (e.g., www.google.com) rather than its IP address.
- The response from the DNS server contains the corresponding IP address.
A Day in the Life: TCP Connection Carrying HTTP
- When a user needs to send an HTTP request, the client opens a TCP socket for that request to the web server.
- TCP transmits a SYN segment and receives a SYNACK segment from the web server, completing the TCP handshake and establishing a connection.
A Day in the Life: HTTP Request/Reply
- The HTTP request is sent into a TCP socket, an IP datagram containing an HTTP request (for www.google.com) is routed to the web server.
- The web server responds with an HTTP reply, including the requested web page, in a TCP packet.
- The response is routed back to the client.
Chapter 5: Summary
- The chapter summarizes the key principles and technologies related to link layer services.
- It covers error detection, error correction, multiple access protocols of link layer addressing, instantiation, and implementation of various link layer technologies, Ethernet, switched LANs, VLANs, and MPLS.
Chapter 5: Let's Take a Breath
- The summary of the 5th chapter touches on journey down the network protocol stack and highlights useful principles and practical knowledge gained in the chapter, except the physical layer (PHY).
- Useful topics for future study could include wireless, multimedia, security, and network management.
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.
Related Documents
Description
This quiz focuses on the link layer of computer networks, covering key concepts such as half-duplex communication, error correction, and MAC address usage. Explore the role of the link layer in both wired and wireless communications and test your understanding of frame transmission and flow control mechanisms.
