π Lecture 64 β A Cross-over Cable versus Leased Circuit
π Overview: This lecture compares an Ethernet cross-over cable, used for direct router-to-router connections, with a leased circuit (WAN link) provided by a telecommunications company. It explains the key differences in transmission speeds, how WAN link speeds are structured, and how to control them using a CSU/DSU device.
ποΈ Topics Covered
The lecture begins by describing how an Ethernet cross-over cable connects two routers for local frame forwarding at speeds like 10 Mbps to 10 Gbps. It then contrasts this with a leased WAN link provided by a telco, which supports many different speeds that are typically multiples of 64 Kbps, such as a T1 line at 1.5 Mbps. Key points about WAN link speed options are outlined, followed by a detailed explanation of how to control the WAN link speed using the channel service unit/data service unit (CSU/DSU) of a serial interface.
π Lecture Summary
What is a Cross-over Cable and a Leased Circuit?
If you install an Ethernet cross-over cable between two routers by plugging the cable into an Ethernet interface on the routers, the two routers can forward Ethernet frames to each other. Ethernet can support transmission speeds of 10 Mbps, 100 Mbps, 1000 Mbps (1 gigabit/second), and even 10 Gbps. If you install a WAN link between two routers, a telco creates something similar to a cross-over cable with some differences. With WAN links, many different transmission speeds are supported. Typically, WAN links use speeds that are multiples of 64 Kbps. For example, a telco offers a service called T1 line or T1 circuit whose speed is 1.5 Mbps, which is 24 times 64 Kbps. A telco can also offer links in multiples of 1.544 Mbps. You can also order even higher speeds that generally come in multiples of 51.84 Mbps.
π‘ Why this matters: This distinction is fundamental for network design. A cross-over cable is a simple, low-cost solution for a direct local connection, while a leased line is a paid service for connecting geographically distant networks with a wide range of speed options.
π Definition β Ethernet cross-over cable: A type of Ethernet cable used to directly connect two similar devices, like two routers, without needing a switch or hub.
π Definition β WAN link: A leased circuit provided by a telecommunications company that connects two routers over a wide geographical area.
π Example: A company in Atlanta and a company in Cincinnati connect their routers via a leased WAN line from a telco. The telco provides a T1 circuit with a speed of 1.5 Mbps, which is equivalent to 24 times the base speed of 64 Kbps.
Key points about speed of WAN links
Many different speeds are possible. You specify the speed when you order the leased line. You need to configure the routers to use the right speed.
How to control the WAN link speed
Serial links run at different speeds, and must be preconfigured. The channel service unit/data service unit (CSU/DSU) of a serial interface controls the speed. The CSU/DSU functions can be done with an external device or as a function of the serial interface card on the router. If you use a CSU/DSU that sits outside the router, you must configure the speed on that external device. If CSU/DSU is built into the router, you must configure the speed of the WAN link on the router.
π Definition β CSU/DSU (Channel Service Unit/Data Service Unit): A device or interface function that controls the speed of a serial WAN link, ensuring the router and the telco's network operate at the same clock rate.
π‘ Why this matters: Without properly configuring the CSU/DSU speed, the serial link will not function, as both ends of the leased line must agree on the transmission clock rate.
β Key Takeaways
A cross-over cable provides a direct Ethernet connection between two routers at speeds like 10, 100, or 1000 Mbps, while a leased WAN circuit is a paid telco service for long-distance connections with speeds typically in multiples of 64 Kbps, such as a T1 at 1.5 Mbps. The speed of a WAN link must be specified when ordering and configured on the routers. The speed of a serial WAN link is controlled by the CSU/DSU, which can be an external device or built into the router's serial interface card. A T1 line's 1.5 Mbps speed is exactly 24 times the base rate of 64 Kbps.
π§ Quick Revision Questions
- What is the primary difference between an Ethernet cross-over cable and a leased WAN circuit in terms of geographical scope and service provider?
- A T1 line is a common WAN service. What is its exact speed, and what is its relationship to the base unit of 64 Kbps?
- Which component of a serial WAN interface is responsible for controlling and clocking the link speed?
- If you have a router with a built-in CSU/DSU, where do you configure the speed of the WAN link?
- What are the possible speeds for an Ethernet cross-over cable, and what are the common speed options for a WAN link as described in the lecture?
π Lecture 65 β Routers and WANS
π Overview: This lecture explains how routers and WANs work together to connect remote sites. It covers the practical steps for installing a WAN link and describes how routers encapsulate IP packets using data link layer protocols like HDLC and PPP for point-to-point connections.
ποΈ Topics Covered
The lecture begins with a concrete WAN link installation plan involving phone company contracts, router and CSU/DSU installation, and cable configuration. It then describes how routers use WAN links to forward IP packets to remote subnets, with end-user devices never connecting directly to the WAN. The lecture concludes with an explanation of WAN data link framing, focusing on HDLC and PPP protocols.
π Lecture Summary
A WAN Link Installation Plan
A step-by-step plan for setting up a leased line WAN connection is provided. First, contract with the phone company to provide a leased line at a certain speed. Second, install a router at each site near where the telco will run its cable. Third, install an external CSU/DSU near the routers at each site, if you didn't buy routers with internal CSU/DSUs. Fourth, configure the CSU/DSUs with the correct speed. Finally, after the phone company runs the cables, install the cables into the CSU/DSU (external) or serial interface of the router (internal CSU/DSU) at each site.
Routers and WANS
Routers use WAN links when they need to forward IP packets to a subnet at a remote site. End user devices at a company's office site use an Ethernet NIC and never directly connect to a WAN link. To send traffic to an IP host at another site, end user devices send the data to the router. The router forwards the IP packet to another router at the other site, which then forwards the packet to the other IP host.
WAN data link framing
When an IP packet arrives to a router, such as R1 acting as Hannah's default gateway, R1 receives the Ethernet frame and checks for errors. If no errors occurred, R1 extracts the IP destination address (e.g., 150.1.3.3) and finds a match in its routing table. R1 then needs to encapsulate the IP packet inside a data link layer frame. Two popular data link layer protocols for point-to-point WAN links are high-level data link control (HDLC) and Point-to-Point Protocol (PPP).
π Definition β HDLC (High-level Data Link Control): A data link layer protocol for point-to-point WAN links that provides framing and error detection.
π Definition β PPP (Point-to-Point Protocol): A data link layer protocol for point-to-point WAN links that provides framing, error detection, and additional features like authentication.
β Key Takeaways
Routers use WAN links to forward IP packets between remote sites, while end user devices connect only via Ethernet and never directly to the WAN. The standard WAN installation process requires a leased line contract, routers, CSU/DSUs (internal or external), and proper cable configuration. When a router receives an Ethernet frame, it extracts the IP destination, checks its routing table, and encapsulates the packet in a WAN data link frame. The two dominant data link layer protocols for point-to-point WAN links are HDLC and PPP, which provide framing and error detection for transmission over the leased line.
π§ Quick Revision Questions
- What are the five steps in the WAN link installation plan?
- Why do end user devices never connect directly to a WAN link?
- What does a router do with an IP packet after it receives an Ethernet frame and finds a match in its routing table?
- What are the two popular data link layer protocols for point-to-point WAN links mentioned in the lecture?
- What is the role of a CSU/DSU in a WAN connection?
π Lecture 66 β Frame Relay
π Overview: This lecture explains Frame Relay, a WAN technology that enables multiple remote sites to communicate using a single physical link per site, reducing hardware and cost compared to leased lines. It covers how Frame Relay works with switches, routers, and virtual circuits, as well as a comparison with HDLC and PPP protocols.
ποΈ Topics Covered
The lecture begins by reviewing HDLC and PPP as point-to-point WAN protocols, noting their single-byte address field and differences. It then introduces Frame Relay as a more efficient WAN technology, explaining its use of Frame Relay switches, virtual circuits, data-link connection identifiers (DLCIs), and how routers communicate across the Frame Relay cloud.
π Lecture Summary
Point-to-Point WAN Protocols (HDLC and PPP)
Point-to-point WAN links use either high-level data link control (HDLC) or Point-to-Point Protocol (PPP) to encapsulate IP packets at the data link layer. Both protocols have only one address field, 1 byte long, but on a point-to-point link, the address doesn't really matter because only the single destination device could receive the data.
π Definition β HDLC: A data link layer protocol defined by the ITU in the 1970s for point-to-point WAN links. π Definition β PPP: A data link layer protocol defined in RFC 1661 during the 1990s with more advanced features than HDLC. π Example: When router R1 sends data to router R2 on a point-to-point WAN link, the HDLC or PPP header contains an address field, but since R2 is the only possible receiver, addressing is trivial.
π‘ Why this matters: Cisco uses a nonstandard version of HDLC but conforms to the standard for PPP.
Frame Relay Basics
Frame Relay is a wide area network (WAN) technology that uses one physical WAN link connected to each site, while allowing each site to send data to every other site. It requires less work, less new hardware, and is more cost-effective than using many serial leased lines.
A Frame Relay network acts like a big WAN switch, with routers connecting to it. To send data to another router, the sending router just sends a frame with the correct address in it.
π Definition β Frame Relay switch: Equipment that understands Frame Relay and can forward traffic based on Frame Relay protocols. The telco uses these switches in its local central offices (COs) to route data to correct sites.
π Example: A company with three remote sites can use Frame Relay: each site has one physical cable to a Frame Relay switch, and data is addressed within the Frame Relay frame to reach any other site, instead of requiring a separate physical link between each pair of sites.
π‘ Why this matters: When a router physically connects to a Frame Relay switch, it connects to a Frame Relay service. The company selling Frame Relay services is called a Frame Relay service provider.
β Key Takeaways
Frame Relay allows multiple remote sites to communicate using a single physical WAN link per site, making it more cost-effective and less hardware-intensive than using multiple leased lines. The network relies on Frame Relay switches in telco central offices to forward frames based on addressing inside the frame. Before Frame Relay, HDLC and PPP were typical for point-to-point links, but they only support two devices per link. Frame Relay essentially acts as a big WAN switch that any connected router can use to reach any other connected router.
π§ Quick Revision Questions
- What is the main advantage of Frame Relay over using multiple leased serial lines to connect multiple sites?
- What is the role of a Frame Relay switch in a wide area network?
- How does the address field in HDLC and PPP differ from addressing in Frame Relay?
- What is a Frame Relay service provider, and what do they provide?
- Why does the address field in an HDLC or PPP frame not matter much on a point-to-point link?
π Lecture 67 β Frame Relay Switching
π Overview: This lecture explains the architecture and operation of Frame Relay, a Layer 2 WAN technology. It covers the physical components, addressing mechanisms, and how Frame Relay switches forward frames between multiple remote sites using DLCIs.
ποΈ Topics Covered
The lecture covers the physical parts of a Frame Relay network, including customer premises equipment (CPE), central office (CO) equipment, and Frame Relay switches. It explains the role of DLCI addresses in frame forwarding, the static configuration requirements for switches, and how the telco's Frame Relay network interconnects multiple remote sites.
π Lecture Summary
Physical Parts of Frame Relay
The Frame Relay network uses the same cabling and CSU/DSU equipment as leased lines, but the telco configures its central office (CO) differently. In the CO, the cable connects to a Frame Relay switch rather than a traditional leased line switch. The Frame Relay provider is typically a telecommunications company (telco) that works with other telcos to create the collective Frame Relay network. The provider's network consists of all its Frame Relay switches and the equipment connecting them. Frame Relay is a set of protocol specifications that operate at OSI Layer 2, the data-link layer.
Frame Relay Switching
When a telco sells Frame Relay service, it forwards Frame Relay frames sent by one router to another router. Before a router can send a packet, it must add the correct data link header and trailer to the packet. Each Frame Relay header contains a single address field called a data-link connection identifier (DLCI). The DLCI is a 10-bit number, usually written as a decimal number between 0 and 1023. Each Frame Relay switch forwards the frame based on the DLCI through the network until it reaches the router on the other side.
π Definition β DLCI (Data-Link Connection Identifier): A 10-bit address field in the Frame Relay header, written as a decimal between 0 and 1023, that identifies the virtual circuit and determines how the frame is forwarded.
π‘ Why this matters: DLCIs are the fundamental addressing mechanism in Frame Relay β they replace the MAC addresses used in Ethernet and determine the path through the provider's network.
Frame Relay switches must be configured statically to know where to forward frames with particular DLCIs in their headers. Unlike Ethernet switches that can learn MAC addresses automatically through flooding and learning algorithms, Frame Relay switches cannot learn addresses automatically. Every switch in the path must have manual configuration entries specifying the correct outgoing interface for each DLCI value.
β Key Takeaways
Frame Relay is a Layer 2 WAN technology that uses DLCI addresses (10-bit values between 0-1023) in the frame header for switching. The physical cabling and CSU/DSU are identical to leased lines, but the telco connects them to Frame Relay switches in the CO. Frame Relay switches forward frames based solely on the DLCI, requiring static configuration of forwarding tables since they cannot learn addresses automatically. The service allows communication between multiple remote sites through the provider's network of interconnected Frame Relay switches.
π§ Quick Revision Questions
- What is the size of the DLCI field in a Frame Relay header, and what range of decimal values can it represent?
- How does a Frame Relay network differ from a leased line network at the central office?
- Why must Frame Relay switches be configured statically rather than learning addresses automatically?
- At which OSI layer does Frame Relay operate, and what is its primary function?
- What equipment must a router use to connect to a Frame Relay network, and is this different from leased line connections?
π Lecture 68 β Virtual Circuits
π Overview: This lecture explains the concept of virtual circuits (VCs) in Frame Relay networks. It demonstrates how a single physical access link can support multiple logical connections to different remote sites, and introduces the distinction between full mesh and partial mesh topologies.
ποΈ Topics Covered
The lecture covers the architecture of Frame Relay networks with multiple remote sites, the definition and characteristics of virtual circuits (VCs), the concept of permanent virtual circuits (PVCs), how service providers preconfigure PVC details using DLCIs, and the difference between full mesh and partial mesh PVC topologies.
π Lecture Summary
Topic 68: Virtual Circuits
In this topic, we explain the concept of virtual circuits.
Let's assume that there are three sites using Frame Relay network and frames are going from one site (R1) to the other two sites (R2 and R3).
In Frame Relay lingo, the ability for R1 to send data to R2 over the Frame Relay network is called a virtual circuit (VC). A VC is like a leased circuit, in that exactly two devices can send and receive data using it. It is called virtual to contrast it with a physical leased circuit. As a VC is typically predefined to always be there, VCs are often called permanent virtual circuits (PVCs).
The service provider preconfigures all the required details of a PVC. Configuring the switches to forward frames with DLCI 102 to Cincinnati and frames with DLCI 103 to Boston. In the previous network, there is no PVC between R2 and R3. The choice of which sites need to have a PVC depends on where the network engineers think that traffic needs to flow in the network.
π Definition β Virtual Circuit (VC): A logical connection between exactly two devices over a Frame Relay network that behaves like a leased circuit but is virtual rather than physical.
π Definition β Permanent Virtual Circuit (PVC): A virtual circuit that is predefined by the service provider to always be available, as opposed to being set up on demand.
π Example: In a three-site Frame Relay network with routers R1 (Atlanta), R2 (Cincinnati), and R3 (Boston):
- A PVC using DLCI 102 connects R1 to R2 (Cincinnati)
- A PVC using DLCI 103 connects R1 to R3 (Boston)
- There is no PVC directly between R2 and R3
- R1 uses its single physical access link to forward traffic to both R2 and R3 simultaneously through these separate PVCs
When routers use Frame Relay, and there is a PVC between each pair of routers, the PVCs are in a full mesh. When not all routers have a PVC, it's called a partial mesh.
π Definition β Full Mesh: A topology where a PVC exists between every pair of routers in the network.
π Definition β Partial Mesh: A topology where not all router pairs have a PVC; only those PVCs that network engineers deem necessary for traffic flow are configured.
π‘ Why this matters: The full mesh vs. partial mesh decision is a critical design consideration. Full mesh provides maximum connectivity but requires more PVCs (and thus more cost), while partial mesh reduces cost but may create traffic bottlenecks or single points of failure.
β Key Takeaways
A virtual circuit (VC) in Frame Relay is a logical connection between exactly two devices that functions like a leased circuit but is virtual rather than physical. VCs that are always present and preconfigured by the service provider are called permanent virtual circuits (PVCs). A single physical access link on a router can support multiple PVCs, each identified by a unique DLCI value, allowing communication with multiple remote sites. When PVCs exist between every pair of routers, the topology is called a full mesh; when some pairs lack direct PVCs, it's a partial mesh. The choice of topology depends on expected traffic patterns and cost considerations.
π§ Quick Revision Questions
- What is a virtual circuit (VC) in the context of Frame Relay, and how does it differ from a physical leased circuit?
- What does PVC stand for, and why are VCs often called PVCs?
- How can a single physical access link on a Frame Relay router support communication with multiple remote sites?
- What is the difference between a full mesh and a partial mesh PVC topology?
- In the three-site example (Atlanta, Cincinnati, Boston), why is there no PVC between R2 and R3, and what determines which PVCs are configured?
π Lecture 69 β Point-to-Point WANs & Frame Relay
π Overview: This lecture compares point-to-point WANs using leased lines with Frame Relay networks. It demonstrates why Frame Relay is considered faster, cheaper, and better for WAN connectivity, focusing on hardware requirements and telco infrastructure differences. Understanding this comparison is crucial for network design decisions regarding cost and efficiency.
ποΈ Topics Covered
This lecture covers the hardware requirements for creating three leased circuits between routers, including serial interfaces and CSU/DSU units. It then presents the equivalent Frame Relay network, highlighting the reduced hardware needs with one cable per router. The lecture visually contrasts the two architectures through figures.
π Lecture Summary
Comparison of Point-to-Point WANs and Frame Relay
Frame Relay is considered to be faster, cheaper, and better than using leased lines for a WAN. Point-to-point WANs tend to require more hardware than does the equivalent network built with Frame Relay.
Three Leased Lines to Connect Three Routers
Figure 69.1 illustrates what the telco does to create three leased circuits to connect three routers. A router requires two serial interfaces and two separate channel service/data service units (CSU/DSU). The telco has to install two cables between the local CO and the office building in Atlanta where R1 resides.
Equivalent Frame Relay Network
Figure 69.2 (implied) shows the equivalent Frame Relay network for the three sites. One cable from each router is needed to the telco, one serial interface, and one CSU/DSU at each router.
β Key Takeaways
Frame Relay requires significantly less hardware than equivalent point-to-point leased line networks. For three sites, each router in a Frame Relay network needs only one serial interface, one CSU/DSU, and one cable to the telco, whereas a point-to-point design requires two serial interfaces, two CSU/DSUs, and two cables per router. This hardware reduction makes Frame Relay faster and cheaper to deploy and maintain.
π§ Quick Revision Questions
- How many serial interfaces does a router need in a three-site point-to-point leased line WAN?
- How many CSU/DSU units are required per router in the equivalent Frame Relay network?
- Why is Frame Relay considered better than leased lines for WANs?
- How many cables must the telco install to the building where R1 resides in the point-to-point scenario?
- What is the main hardware difference between the two approaches for connecting three routers?
π Lecture 70 β Routing over PVCs
π Overview: This lecture explains how routing works over Permanent Virtual Circuits (PVCs) in a Frame Relay network. It addresses the challenge of connecting multiple sites efficiently without requiring separate physical interfaces for each remote connection, and introduces key concepts like partial mesh topologies and data-link connection identifiers (DLCIs).
ποΈ Topics Covered
This lecture covers how to configure routing over PVCs in a Frame Relay network with three sites, using a partial mesh topology where not all routers have direct PVCs to each other. It explains the role of data-link connection identifiers (DLCIs) in identifying PVCs and how routers use them to forward frames between sites. The lecture also discusses the cost and hardware implications of scaling point-to-point WANs versus using PVCs.
π Lecture Summary
Routing over PVCs
This topic explains routing over permanent virtual circuits (PVCs). In a growing company with multiple sites, using point-to-point WANs becomes impractical because each additional point-to-point line requires a separate physical serial interface and a separate CSU/DSU on the router. This creates significant extra costs for router hardware, CSU/DSU equipment, and the telco must run many more cables.
Assume an internetwork that uses three routers at three sites. Each router has an access link to a local Central Office (CO). In this design, there is no PVC between R2 and R3, making the network a partial mesh of PVCs. The Frame Relay provider (the telco) tells the customer to configure R1 so that it uses data-link connection identifier (DLCI) 102 to send frames to R2, and DLCI 103 to send frames to R3.
π Definition β PVC (Permanent Virtual Circuit): A logical, permanent connection between two endpoints in a Frame Relay network that functions like a dedicated line but uses the provider's shared infrastructure.
π Definition β DLCI (Data-Link Connection Identifier): A 10-bit address field in the Frame Relay frame header that identifies which PVC a frame should traverse, essentially functioning as the "to" address for Frame Relay frames.
π Formula: PVC mapping β DLCI value at router interface maps to specific remote destination
- Meaning: Each local DLCI tells the router which PVC to use to reach a specific remote router
π Example: In Figure 70.1, a three-router Frame Relay network has:
- R1 configured with DLCI 102 β used to send frames to R2
- R1 configured with DLCI 103 β used to send frames to R3
- No PVC exists between R2 and R3 (partial mesh)
- Each router connects to the local CO via an access link
π‘ Why this matters: Using PVCs allows a router to reach multiple remote sites using a single physical serial interface and CSU/DSU, dramatically reducing hardware costs compared to point-to-point lines which require one physical interface per remote connection.
β Key Takeaways
The fundamental difference between point-to-point WANs and PVCs is that each point-to-point line requires separate physical interfaces and CSU/DSU equipment, while PVCs allow one physical interface to support multiple logical connections using DLCIs. In a partial mesh Frame Relay network, not all routers have direct PVCs between them (as with R2 and R3 in the example). DLCI values are assigned by the Frame Relay provider and tell the router which PVC to use for each destination. Scaling a point-to-point WAN from 3 to 100 sites requires significantly more hardware and cabling, making PVCs a more cost-effective solution.
π§ Quick Revision Questions
- What is the difference between a point-to-point WAN and using PVCs in terms of hardware requirements?
- What does DLCI stand for and what is its purpose in a Frame Relay network?
- What is a partial mesh topology and how does it differ from a full mesh?
- Why would a company choose to use PVCs rather than point-to-point lines as it grows from 3 sites to 10 sites?
- In the example with R1, R2, and R3, why doesn't R1 use a single DLCI to reach both R2 and R3?
π Lecture 71 β The Internet: A Large IP Network
π Overview: This lecture explains how the Internet enables global IP packet delivery through a single connection. It contrasts the Internet with WAN links and Frame Relay, showing how an Internet service provider (ISP) creates an IP network that allows enterprises and individuals to communicate worldwide. The lecture also reviews the three major components that make up the Internet.
ποΈ Topics Covered
This lecture begins by explaining how a company can communicate between far-away sites using telco services, including WAN links and Frame Relay. It then shows how a single connection to the Internet provides access to a global network, using an ISPβs IP network. Finally, it presents a comparison table of WAN links, Frame Relay, and the Internet, summarizing their key differences and capabilities.
π Lecture Summary
The Internet: A Large IP Network
This topic explains why connecting to the Internet provides a path for IP packets to be sent almost anywhere. A company cannot run a cable between two far-away buildings for legal and practical reasons. A telco already has the ability to let you connect to its network physically, and then deliver the data over its network. With WAN links, the telco allows two sites to communicate. With Frame Relay, many sites can communicate, but they all must use the same Frame Relay service from the same telco. The company can connect to the Internet with a WAN link or Frame Relay. With that one physical connection, that company gains access to a huge number of people who also have access to the Internet.
π‘ Why this matters: The Internetβs single-connection model drastically reduces the cost and complexity of connecting to a global network, unlike WAN links or Frame Relay which are limited to private, telco-specific networks.
Reaching the World through One Connection to the Internet
An Internet service provider (ISP) creates an IP network over which IP packets can be forwarded, similar to the way a telco creates a network over which bits (for WAN links) and Frame Relay frames (for Frame Relay) can be forwarded. Some ISPs are also telcos. The Internet consists of three major components:
- The IP networks that ISPs create.
- The enterprise IP networks that attach to one or more ISPs.
- The individuals who connect their computers to ISPs.
π Definition β ISP (Internet Service Provider): A company that creates an IP network to forward IP packets, enabling customers to access the Internet.
π Concept: A single physical connection from a company to an ISP provides access to a huge number of other people and organizations also connected to the Internet.
π Example: If a company connects to an ISP using a single WAN link or Frame Relay circuit, that one connection allows IP packets to reach any other device on the Internet, rather than requiring separate links to each destination.
Comparisons of WAN Links, Frame Relay and the Internet
Table 71.1 compares the three technologies across several dimensions:
- Number of sites: WAN links connect two sites; Frame Relay connects many sites; the Internet connects many sites.
- Structure: WAN links require a full mesh of links for many-to-many communication; Frame Relay uses a partial mesh of PVCs; the Internet uses a partial mesh of connections to ISPs.
- Connection type: WAN links use point-to-point circuits; Frame Relay uses PVCs; the Internet uses any WAN technology to connect to the ISP.
- Addressing: WAN links use no extra addressing (just the circuit); Frame Relay uses DLCI identifiers; the Internet uses IP addresses.
- Worldwide reach: WAN links do not provide worldwide reach; Frame Relay does not provide worldwide reach; the Internet does provide worldwide reach.
π Definition β Worldwide reach: The ability to send data to any destination on the globe, which only the Internet provides among these three technologies.
π Table: The key differentiator is that only the Internet offers worldwide reach, while WAN links and Frame Relay are confined to the telcoβs private network.
β Key Takeaways
The Internet enables global IP packet delivery through a single connection to an ISP, unlike WAN links which only connect two sites and Frame Relay which connects multiple sites but within the same telcoβs network. The Internet comprises three major components: ISP IP networks, enterprise IP networks, and individual usersβ computers. WAN links and Frame Relay are private telco services limited to subscribers of that specific telco, while the Internetβs structure allows any device with an IP address to communicate worldwide. The comparison table highlights that only the Internet offers worldwide reach, with IP addressing being the universal addressing scheme.
π§ Quick Revision Questions
- What are the three major components that make up the Internet?
- How does a single connection to the Internet provide access to a global network, while WAN links and Frame Relay do not?
- What is an ISP, and how does it differ from a telco in terms of the network it creates?
- According to Table 71.1, which technology (WAN links, Frame Relay, or the Internet) allows for many sites to communicate and also provides worldwide reach?
- Why canβt a company simply run a cable between two far-away buildings, and how do telcos and ISPs solve this problem?
π Lecture 72 β Using a Phone Line for Data
π Overview: This lecture explains how modems enable data transmission over standard analog telephone lines. It describes the fundamental characteristics of analog signals (frequency and amplitude) and how modems convert digital computer data into analog signals for transmission over the Public Switched Telephone Network (PSTN), mimicking voice signals without requiring physical changes to the local loop.
ποΈ Topics Covered
The lecture covers the basic components of a telephone system (microphone, local loop, telco CO, speaker), the characteristics of analog electrical signals (frequency and amplitude), how voice works as a continuously changing analog wave, and how modems convert binary digits from computers into representative analog signals for transmission over the PSTN.
π Lecture Summary
Using a Phone Line for Data
A telephone includes a microphone, which converts sound waves into an analog electrical signal. This signal is carried by the local loop (the phone line between a house and a telco CO). The telco sends the electrical signal between one phone and another. On the receiving side, the phone converts the electrical signal back to sound waves using a speaker. The characteristics of the analog signal are:
- Frequency: the number of times a signal repeats itself, from peak to peak, in one second.
- Amplitude: how strong the signal is.
π Definition β Local loop: The phone line between a house and a telco Central Office (CO). π Definition β Frequency: The number of times a signal repeats itself, from peak to peak, in one second. π Definition β Amplitude: How strong the signal is. π Formula: Frequency = number of cycles per second (measured in Hertz, Hz) β the higher the frequency, the more cycles per second. π Example: An analog electrical signal with a high amplitude would represent a louder sound wave compared to one with a low amplitude, for the same frequency.
The sounds that your voice makes happen to be a continuously changing sound wave, so analog electrical signals that continuously change work well for voice. Modems allow two computers to send and receive a serial stream of bits over an analog phone circuit, with no physical changes required on the local loop between a residence and the telco CO. As the telephone switch in the CO expects to send and receive analog voice signals over the local loop, modems simply send an analog signal to the PSTN and expect to receive an analog signal from the PSTN. Instead of voice, a modem converts a string of binary digits on a computer into a representative analog electrical signal.
π Definition β Modem: A device that allows two computers to send and receive a serial stream of bits over an analog phone circuit. π Definition β PSTN: Public Switched Telephone Network. π‘ Why this matters: Modems enable data communication over the existing telephone infrastructure without requiring any physical modifications to the local loop, which was critical for early internet and remote access.
β Key Takeaways
Analog telephone signals have two key characteristics: frequency (how often the signal repeats per second) and amplitude (signal strength). Voice is a continuously changing analog wave, so analog signals work well for it. Modems enable computers to transmit digital data (binary digits) over analog phone lines by converting the data into representative analog electrical signals, effectively mimicking voice. This entire process works without physical changes to the local loop between the residence and the telco CO, as the telephone switch expects analog signals. A modem on the sending side converts bits to analog, and on the receiving side, a modem converts the analog signal back to bits.
π§ Quick Revision Questions
- What are the two key characteristics of an analog electrical signal mentioned in the lecture?
- What does a microphone do in a telephone system?
- How does a modem allow two computers to communicate over an analog phone circuit?
- Why can modems use the existing local loop without physical changes?
- What does a modem convert on the sending side, and what does it expect to receive from the PSTN?
π Lecture 73 β Digital Subscriber Line (DSL)
π Overview: This lecture introduces Digital Subscriber Line (DSL) technology as an alternative to dial-up modems for Internet access. It explains how DSL overcomes the limitations of traditional modems by providing much faster speeds while allowing simultaneous phone calls and Internet usage over the same telephone line.
ποΈ Topics Covered
The lecture compares modems to DSL technology, explaining how modems encode binary data onto analog signals through modulation and are limited by speed and inability to use the phone simultaneously. It then introduces DSL as a superior alternative that uses the same phone line but operates at much faster speeds while enabling concurrent voice calls and Internet surfing.
π Lecture Summary
Modems and Their Limitations
Modems encode a binary 0 or 1 onto the analog signal by varying frequency or amplitude. Changing these characteristics of the analog signal is referred to as modulation. In the previous example, frequency changes every 0.1 seconds, so a PC can send 10 bits/sec. If a modem sends at 9600 bps, the receiving modem needs to sample the incoming analog signal every 1/9600 of a second. The details of modem operation relate to OSI Layer 1, the physical layer.
Modems use the phone line to make what the telco thinks is a voice call, so you cannot make another call while you are surfing the Internet. Another big negative is the speed.
Digital Subscriber Line (DSL)
Digital subscriber line (DSL) is an alternative technology for sending and receiving data to and from an ISP, using the same old phone line, but running at much faster speeds. Also, you can make phone calls and surf (send IP packets) at the same time.
π Definition β Digital Subscriber Line (DSL): A technology that uses existing telephone lines to provide high-speed Internet access while simultaneously allowing voice calls.
π‘ Why this matters: DSL revolutionized home Internet access by eliminating the need to tie up the phone line for web surfing, while offering significantly higher speeds than dial-up modems.
β Key Takeaways
DSL is a major improvement over traditional modems because it allows simultaneous voice calls and Internet usage on the same phone line. While modems require the phone line to be dedicated to a single voice call during Internet use and are limited to speeds like 9600 bps, DSL provides much faster data transmission. Both technologies operate at the physical layer (OSI Layer 1) but use different methods to encode data onto analog signals. Modulationβchanging frequency or amplitudeβis the fundamental technique modems use to represent binary data.
π§ Quick Revision Questions
- What two characteristics of an analog signal can be varied to encode binary data in modulation?
- At what speed does a modem send if the receiving modem must sample every 1/9600 of a second?
- What OSI layer do modems and DSL operate at?
- What is one major limitation of modems that DSL overcomes regarding phone usage?
- What does DSL stand for and what is its primary advantage over traditional modems?
π Lecture 74 β Sending Data without a Phone Line
π Overview: This lecture describes how existing cable television infrastructure can be used for Internet access, bypassing the need for a traditional phone line. It explains the hybrid fiber-coaxial (HFC) access network, its components, and how frequency division multiplexing enables data transmission over cable TV networks.
ποΈ Topics Covered
This lecture covers how cable Internet access leverages the existing cable television infrastructure, from the cable head end to individual residences. It introduces the concept of a hybrid fiber-coaxial (HFC) network, where fiber optics connect the head end to neighborhood junctions, and coaxial cable reaches individual houses. The lecture also explains frequency division multiplexing as the method for transmitting different channels in different frequency bands.
π Lecture Summary
Sending Data without a Phone Line
This topic describes how TV cable networks can be used for the Internet. Cable Internet access makes use of the cable television companyβs existing cable television infrastructure. A residence obtains cable Internet access from the same company that provides its cable television. A cable head end broadcasts television channels.
Hybrid Fiber-Coaxial (HFC) Access Network
To connect houses to the Internet, a hybrid fiber-coaxial (HFC) access network is used. Fiber optics connects the cable head end to neighborhood-level junctions, from which traditional coaxial cable is then used to reach individual houses. This architecture is called Hybrid fiber coax (HFC).
Frequency Division Multiplexing
The lecture introduces frequency division multiplexing as the method used in these networks. This technique allows different channels to be transmitted simultaneously in different frequency bands without interference.
β Key Takeaways
Cable Internet access uses the existing cable television infrastructure, not phone lines. A hybrid fiber-coaxial (HFC) network combines fiber optics from the cable head end to neighborhood junctions with traditional coaxial cable to individual houses. Frequency division multiplexing is the key technique that enables transmitting different channels in different frequency bands, allowing both television and Internet data to coexist. The cable head end is the central point that broadcasts television channels and connects to the Internet backbone.
π§ Quick Revision Questions
- What is the central point in a cable television network that broadcasts television channels?
- What type of cable connects the cable head end to neighborhood-level junctions in an HFC network?
- What type of cable is used from neighborhood junctions to reach individual houses?
- What is the name of the system that combines fiber optics and coaxial cable for Internet access?
- What technique allows different channels to be transmitted in different frequency bands simultaneously?
π Lecture 75 β Introduction to AAA Security Model
π Overview: This lecture introduces the AAA security model, which stands for Authentication, Authorization, and Accounting. Understanding these three components is critical for controlling access to servers and networks, ensuring that only authorized users can perform specific actions.
ποΈ Topics Covered
The lecture defines AAA and explains its three components. It details authentication as a five-step process using a username and password, describes how an authentication server works with application servers using the Kerberos protocol, and defines authorization as determining what a user is allowed to do. The accounting component is mentioned but not detailed.
π Lecture Summary
AAA Stands for Authentication, Authorization, and Accounting
AAA refers to three areas of security related to an individual device or user. Before using most servers, you must have the right to use that server, typically using a username and password.
π Definition β AAA: A security model that stands for Authentication, Authorization, and Accounting, addressing issues regarding individual device or user access rights.
Authentication
Authentication is a five-step process:
- Fred requests a web page.
- The web server replies, asking for a username and password.
- Fred types in and sends his username and password to the server.
- The web server checks its list of usernames and verifies that the password is correct.
- If the username and password are correct, the web server returns the contents of the web page.
Username and password can be listed on a server called an authentication server. Application servers query the authentication server. The Kerberos protocol runs between application and authentication servers.
π Definition β Authentication: The process of verifying the identity of a user or device, typically using a username and password.
π‘ Why this matters: Authentication is the first line of defense in controlling access to network resources and systems.
Authorization
The second "A" in AAA is for Authorization. It refers to the process of figuring out what a particular user is allowed to do.
π Definition β Authorization: The process of determining what actions or resources a verified (authenticated) user is permitted to access.
β Key Takeaways
AAA stands for authentication, authorization, and accounting, forming a comprehensive security model for controlling user access. Authentication is a five-step process where a user proves their identity using credentials like a username and password. Authorization determines the specific permissions and actions a user is allowed to perform after being authenticated. Authentication servers and protocols like Kerberos facilitate secure verification between application servers and user credentials.
π§ Quick Revision Questions
- What are the three components of the AAA security model?
- List the five steps in the basic authentication process.
- What is the role of an authentication server in the AAA model?
- Which protocol runs between an application server and an authentication server?
- What is the difference between authentication and authorization in the AAA model?