Untitled Quiz
28 Questions
0 Views

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to lesson

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What are the two approaches to structuring network control plane?

  • Link State and Distance Vector
  • Per-router control and logically centralized control (correct)
  • Data Plane and Control Plane
  • Routing Algorithm and Forwarding

    • What is the goal of a routing protocol?

    To determine the “good” paths (equivalently, routes), from sending hosts to receiving host, through a network of routers

    What are the two main categories of routing algorithms classified based on the rate of changes in routes?

  • Static and Dynamic (correct)
  • Link State and Distance Vector
  • BGP and OSPF
  • Centralized and Decentralized

    • Link_state_ routing algorithms are decentralized.

    <p>False (B)</p> Signup and view all the answers

    In the context of routing, what does a "good" path refer to?

    <p>The path with the least cost, fastest, and least congested (C)</p> Signup and view all the answers

    What is the essential idea behind the Distance Vector algorithm?

    <p>Each node sends its own distance vector estimate to its neighbours, and when they receive new distance vector estimates from any neighbors, they update their own distance vector using the Bellman-Ford equation.</p> Signup and view all the answers

    The Bellman-Ford equation is used in Distance Vector routing algorithm.

    <p>True (A)</p> Signup and view all the answers

    In the context of routing, what is the "count-to-infinity" problem?

    <p>It occurs when a link cost changes, the Distance Vector algorithm can experience a delay in updating the routes. When a router detects a change, it starts to propagate this information via distance vector updates. However, these updates can circulate in the network multiple times, resulting in a prolonged period of incorrect routes.</p> Signup and view all the answers

    Which of these is a disadvantage of the Distance Vector algorithm?

    <p>It can be slow to converge to the correct routes (D)</p> Signup and view all the answers

    What are the two main techniques used to address the issue of scalability in internet routing?

    <p>Hierarchical Routing and Policy-Based Routing (B)</p> Signup and view all the answers

    What is the purpose of hierarchical routing in Internet?

    <p>It divides the internet into autonomous systems (AS), with a dedicated routing protocol for each AS, and another protocol for routing between these ASes. This approach helps reduce the size of routing tables and the frequency of updates needed.</p> Signup and view all the answers

    Which of these is a common intra-AS routing protocol?

    <p>RIP (B), OSPF (D)</p> Signup and view all the answers

    The BGP protocol is designed for intra-AS routing.

    <p>False (B)</p> Signup and view all the answers

    What is the main purpose of eBGP?

    <p>It is used to exchange subnet reachability information between neighboring ASes (Autonomous Systems) to establish connections between different internet providers or companies.</p> Signup and view all the answers

    What does the AS-PATH attribute indicate within the BGP routing protocol?

    <p>The list of ASes the prefix advertisement has passed through (D)</p> Signup and view all the answers

    What is the "hot potato" routing principle in BGP?

    <p>The principle dictates that a router chooses the local gateway with the least intra-domain cost, even if it involves more AS hops, to prioritize minimizing cost within the AS.</p> Signup and view all the answers

    SDN is a logically decentralized control plane.

    <p>False (B)</p> Signup and view all the answers

    What is the primary benefit of SDN compared to traditional network control?

    <p>SDN enables easier network management by centralizing control and allowing for simple configuration adjustments, thereby offering greater flexibility in traffic routing compared to the traditional per-router approach.</p> Signup and view all the answers

    Which of the following is a potential advantage of SDN from a programmability perspective?

    <p>It enables creating custom routing policies and new network functionalities (D)</p> Signup and view all the answers

    OpenFlow is a southbound API.

    <p>True (A)</p> Signup and view all the answers

    What is the main purpose of the OpenFlow protocol?

    <p>OpenFlow enables a central controller to communicate with and program switches in the data plane to make routing decisions based on flow rules that are managed by the controller. This offers a fundamental level of interaction between the control and data planes within SDN.</p> Signup and view all the answers

    What is the main function of ICMP (Internet Control Message Protocol)?

    <p>To allow devices to communicate network-level information (A)</p> Signup and view all the answers

    ICMP is a transport layer protocol.

    <p>False (B)</p> Signup and view all the answers

    Which of the following is a common network management protocol?

    <p>SNMP (C)</p> Signup and view all the answers

    What is a key advantage of using NETCONF over SNMP for network management?

    <p>NETCONF offers a richer set of capabilities compared to SNMP. It supports more complex actions like retrieving and modifying configuration data, allowing for more detailed network management and automation features. It also utilizes a more robust and secure communication mechanism using XML and transport protocols like TLS for enhanced security and reliable communication.</p> Signup and view all the answers

    YANG is a standardized data modeling language used in NETCONF.

    <p>True (A)</p> Signup and view all the answers

    What is the purpose of the YANG protocol?

    <p>YANG is designed to define data models and structures for network management, particularly within the context of NETCONF. It ensures consistency, correctness, and predictability in how data is formatted and managed within the network.</p> Signup and view all the answers

    What does the term "atomic commit" refer to in the context of NETCONF?

    <p>A group of changes that are applied to multiple devices in a single transaction (C)</p> Signup and view all the answers

    Study Notes

    Network Layer Control Plane

    • PowerPoint slides are freely available for faculty, students, and readers.
    • Users can modify, add, and delete slides as needed.
    • Slides are used in classes.
    • If slides are posted online, it's necessary to cite their origin and copyright.

    Network Layer Control Plane: Goals

    • Understanding the principles behind the network control plane
    • Traditional routing algorithms
    • SDN (software-defined networking) controllers
    • Network management and configuration
    • Instantiation and implementation in the Internet
    • OSPF (Open Shortest Path First), BGP (Border Gateway Protocol)
    • OpenFlow, ODL, and ONOS controllers
    • Internet Control Message Protocol (ICMP)
    • SNMP (Simple Network Management Protocol)
    • YANG/NETCONF

    Network Layer: "Control Plane" Roadmap

    • Introduction
    • Routing Protocols (link-state, distance vector)
    • Intra-ISP routing (OSPF)
    • Routing among ISPs (BGP)
    • SDN control plane
    • Internet Control Message Protocol (ICMP)
    • Network management and configuration (SNMP, NETCONF/YANG)

    Network-Layer Functions

    • Forwarding: move packets from input to appropriate router output
    • Routing: determine the path taken by a packet from source to destination
    • Two approaches to structuring network control plane:
      • Per-router control (traditional)
      • Logically centralized control (software-defined networking)

    Per-Router Control Plane

    • Individual routing algorithm components exist in each router.
    • These components interact within the control plane.
    • Components compute forwarding tables.
    • Forwarding tables include header values and output values.

    Software-Defined Networking (SDN) Control Plane

    • Remote controller computes and installs forwarding tables in routers.
    • The controller manages forwarding tables.
    • Values from arriving packets are used by the controller.
    • The controller has control over the data plane.

    Routing Protocols

    • Routing protocols aim to determine good paths (routes) for data packets from a source host to a destination host.
    • The paths are through a network of routers.
    • The goal is to find the shortest, fastest, or least-congested paths.
    • G = (N, E) describes a graph with a set of routers (N) and a set of links (E).
    • Links connect pairs of routers.
    • Link costs (e.g., cw,z = 5, Cu,z) depend on the network operator.
    • Link costs might correspond inversely to bandwidth.

    Routing Algorithm Classification

    • Static routes change slowly over time. All routers have complete topology and link cost information
    • Dynamic routes change more quickly. Routers exchange information periodically or in response to link cost changes.
    • Decentralized mechanisms iterate to compute and share routing information.
    • Routers initially know only link costs to attached neighbors.
    • Distance vector algorithms are global or decentralized.
    • Centralized network topology, all nodes know link costs (using "link-state broadcast").
    • All nodes share the same information.
    • Computes least cost paths from a source node to all other nodes.
    • Iterative process to determine the least cost path for each destination.

    Dijkstra's Algorithm: An Example

    • Provides steps and the process of how Dijkstra's algorithm works.

    Dijkstra's Algorithm: Another Example

    • Illustrates another example of the Dijkstra algorithm

    Dijkstra's Algorithm: Discussion

    • Algorithm complexity of the Dijkstra algorithm:
      • O(n²) complexity per iteration for n nodes.
      • O(n log n) efficiency is achievable in optimized implementations.
    • Message complexity of Dijkstra algorithm: O(n²)
    • Each router must broadcast link state info to all other n routers, which takes overall O(n) link crossings.

    Dijkstra's Algorithm: Oscillations Possible

    • Link costs depend on traffic volume. Route oscillations are possible, depending on costs involved.
    • Traffic entering at multiple locations (e.g., d, c, e with rates 1, e <1) creates situations where link costs are directional, and volume-dependent.
    • Dynamic adjustments to costs to result in changes in routes.

    Distance Vector Algorithm

    • Time-based estimation of hop count distances
    • Each node sends its current distance estimations to its neighbors.
    • Receives new DV estimates from neighbor. Updates it's own DV information.
    • The distance vector algorithm calculates minimal cost paths.

    Distance Vector Algorithm: Key Idea

    • Iterative asynchronous approach, each node sends distance vector estimates to its neighbors.
    • Neighbor receiving new estimates updates its own distances.
    • Values converge to the least cost under normal conditions.

    Distance Vector Algorithm: Each Node

    • Waits for changes in local link costs or messages from neighbors.
    • Recomputes and updates DV estimations.
    • Notifies neighbors if DV changes.

    Distance Vector Algorithm: Example

    • Illustrates how the algorithm works in an example.

    Distance Vector Algorithm: Iteration

    • Nodes receive distance vectors from neighbors.
    • Nodes compute their new local distance vector.
    • Send new local distance vector to neighbors.

    Distance Vector Example: Computation

    • Iterative process of distance vector calculation based on initial vectors by each node.

    Distance Vector: State Information Diffusion

    • Iterative communication where computation steps are distributed across a network based on state information.
    • If a link cost changes, the node updates its routing information and recalculates its local DV.
    • If needed, notify the relevant neighbors.
    • If a link cost changes, the node updates its DV. The node informs its neighbors, and the neighbors update their tables.
    • No new cost, so it sends no further messages.

    BGP (Border Gateway Protocol)

    • Routing protocol used across domains ('autonomous systems').
    • Allows subnets to advertise their existence and reachability.
    • Provides means for ASes to obtain subnet reachability information from neighboring ASes and to pass information to all its internal routers.
    • Determines good routes between networks.

    eBGP, iBGP Connections

    • eBGP: Connections between gateway routers occur between different autonomous systems (ASes).
    • iBGP: Connections occur within the same autonomous system (AS).
    • Router gateways perform both eBGP and iBGP protocols.

    BGP Basics

    • BGP routers exchange messages using TCP connections.
    • Each message advertises paths to different networks.

    Path Attributes and BGP Routes

    • BGP advertisements contain prefixes (destination network IDs) and attributes.
    • AS path identifies all the autonomous systems traversed by the route.
    • Next-hop specifies the outgoing interface for the next router along the path.
    • Policy-based routing influences the router's import policies for handling paths advertised from other ASes.

    BGP Path Advertisement

    • AS2 router receives path advertisement.
    • AS2 router propagates the path advertisement to other routers in AS2 based on policy.
    • AS2 router advertises path to AS1 based on configuration policy.

    BGP Path Advertisement (More)

    • Routes learned from different paths exist.
    • An AS1 gateway router can learn multiple paths.
    • Policies influence the chosen path.

    BGP Messages

    • BGP exchanges messages over TCP connections.
    • OPEN message initiates the connection.
    • UPDATE messages advertise/withdraw paths. KEEPALIVE messages maintain connections.
    • NOTIFICATION messages report errors.

    BGP Path Advertisement

    • Routers learn paths via iBGP from neighboring ASes.
    • Internal intradomain routing is used to discover the specific interface.

    Why Different Intra-, Inter-AS Routing?

    • Policy differences exist in managing traffic control within and between Autonomous Systems (ASes).
    • The concept of administrative autonomy has influenced decisions made in routing.
    • Policies and scalability demands may vary. Intra-AS routing uses simpler, less complex procedures to suit its purposes, while different protocols, including those in inter-AS, are designed to be robust, secure, and address different routing concerns.

    Hot Potato Routing

    • Routers choose local gateway routes that have the lowest intradomain cost.
    • Routers do not necessarily concern themselves with overall interdomain costs.

    BGP: Achieving Policy via Advertisements

    • ISPs might not want to carry transit traffic between other ISPs.
    • A network typically includes provider networks and customer networks.
    • Policies are implemented, as provider networks sometimes do not want to exchange traffic with other providers (or other customer networks).

    BGP: Achieving Policy via Advertisements (More)

    • ISPs frequently want to keep their transit traffic to their customer networks only. Routing between provider networks often follows policies.

    BGP Route Selection

    • Routers might learn various paths to a destination.
    • Route selection criteria include:
      • Local preference value
      • Shortest AS path
      • Closest next hop
    • Additional criteria might also be used.

    Network Layer: Summary

    • Review of the concepts discussed in the lecture notes, such as the approaches to network control plane, traditional routing, SDN controllers, protocols, and so on.

    SDN: Selected Challenges

    • Hardening the control plane for reliable performance and robustness to failures.
    • Ensuring dependability and security as well as protocols that meet mission specific requirements.
    • Scaling internet routing beyond a single autonomous system (AS).

    SDN and the Future of Traditional Network Protocols

    • SDN-computed versus router-computed forwarding tables.
    • Example of logical centralization of computed protocols, or methods such as congestion control based on router-reported congestion data.

    Network Layer: Roadmap

    • Summary of the topics and concepts studied in network layer control plane.

    ICMP: Internet Control Message Protocol

    • Hosts and routers use ICMP for error reporting and other network level information.
    • The ICMP protocols employ messages to communicate host and network level issues (e.g., unreachable hosts or networks).

    Traceroute and ICMP

    • Source sends UDP segments to a destination.
    • TTL is incrementally decreased to identify routers along the path.

    Additional Chapter 5 Slides

    • Notes for additional chapter 5 slides

    Distance Vector: Another Example

    • Illustrates the distance vector calculation in an example.

    Distance Vector: Another Example

    • Illustrates another example of the distance vector computation.

    Network Management

    • Autonomous systems are made up of thousands of interacting hardware and software components.
    • Network management is the process of integrating these components.
    • Managing these components, which can include monitoring and configuration, requires sophisticated coordination and control.

    Components of Network Management

    • Network management server/controller manages the data.
    • Managed devices, such as computers and routers, interact with the management server/controller.

    Network Operator Approaches to Management

    • CLI (command-line interface) offers direct communication with network devices.
    • SNMP (Simple Network Management Protocol) facilitates data transfer among network management devices.
    • Netconf/Yang is a more abstract approach to network management, focusing on holistically managing data within a larger network scope.

    SNMP Protocol

    • Two modes of conveying MIB information and commands to/from the network management server are in use:
      • Request/response mode: Messages are exchanged with explicit requests and responses.
      • Trap mode: Asynchronous communication to carry notifications about significant or exceptional events.

    SNMP Protocol: Message Types

    • Various message types are used to convey different commands or information between the management server and the managed devices.

    SNMP Protocol: Message Formats

    • Message header elements, such as the type of request, error status, indexes, the name value pairs, and so on, are conveyed between different entities in the network to carry out the commands and inform the relevant network entities or component, if needed.

    SNMP: Management Information Base (MIB)

    • MIB modules gather operational and configuration information of networked devices.
    • SMI defines the structures of MIBs and offers data definition language for creating MIBs.
    • Example MIB variables for UDP protocol, which include information like the total number of delivered or undeliverable data segments, are included in the protocols for managing networked systems efficiently.

    NETCONF Overview

    • Actively manages and configures network devices.
    • Operates between the managing server and managed devices.
    • Used for configurations.
    • Supports atomic operations to allow consistent modification of multiple devices in the network.

    NETCONF Initialization, Exchange, Close

    • Protocol messages exchanged during protocol initialization and closing operations. This section illustrates message formats.

    Selected NETCONF Operations

    • NETCONF operations and their descriptions.
    • Descriptions that explain each of the operations include explanations of configuration states and operational state, managing data through commands to alter configurations, or actions that can be taken for particular network components.

    Sample NETCONF RPC Message

    • Examples of NETCONF RPC messages.

    YANG Data Modeling Language

    • YANG is a data modeling language for configuring network devices that incorporates data types, structure, and semantics for describing the network. YANG descriptions are written to match and satisfy correctness and consistency constraints.

    Network Layer Summary

    • A summary of the topics covered in the network layer discussions or lectures.

    OpenFlow Protocol

    • Exchange messages between controller and switch using TCP
    • Supports optional encryption
    • Three types of message: controller-to-switch, synchronous (switch to controller), and symmetric (misc) messages.

    OpenFlow: Controller-to-Switch Messages

    • Features queries,replies about switch capabilities to the controller.
    • Configure switch configuration parameters.

    OpenFlow: Switch-to-Controller Messages

    • Packet-in messages transfer packet data to the controller.
    • Flow-removed messages remove flow entries from the switch's flow table.
    • Port status informs the controller about any changes in port status.

    SDN: Control/Data Plane Interaction Example

    • Controller interacts with the network layer to handle link failures.
    • The SDN controller updates flow tables and computes new routes in response to detected link failures.

    OpenDaylight (ODL) Controller

    • REST, RESTCONF, and NETCONF northbound and southbound APIs
    • Enhanced Services for managing and monitoring various network components

    ONOS Controller

    • Control apps separate from the controller.
    • Emphasizes distributed core operations.
    • Provides high-level specification using an intent framework.

    SDN: Selected Challenges

    • Challenges associated with hardening the control plane, robustness in the face of failures, distributed system security, and scaling internet routing.

    SDN and the Future of Traditional Network Protocols

    • Comparison of SDN-computed forwarding tables and router- computed methods.
    • SDN can compute congestion levels, allowing for flexible sender rate control.

    Network Layer: Roadmap

    • Summary of the topics and concepts learned in the network Layer, pertaining to its control plane.

    ICMP: Internet Control Message Protocol

    • Function of ICMP and how it's used in network-level information communication.
    • ICMP messages inform network-level issues.

    Traceroute and ICMP

    • Traceroute uses UDP segments to locate destination routers via ICMP feedback from various intermediary routers.

    Internet Control Message Protocol (ICMP) Messages

    • Includes types, codes, and the first 8 bytes of error-inducing IP datagram payload.

    Network Management

    • Description of how network management systems are used to monitor, configure, and control network devices.
    • Detailed description of roles of network management systems.

    Components of Network Management

    • Describes the different components used to manage network devices (e.g, the management server, managed devices, agents, and the protocols they use).

    Network Operator Approaches to Management

    • Describes various management approaches, from the CLI (command-line interface) to SNMP (Simple Network Management Protocol), and other management protocols (e.g., NETCONF and YANG).

    SNMP Protocol (Two modes: Request/Response & Trap)

    • Describes how SNMP messages are conveyed in two modes.

    SNMP Protocol: Message Types

    • Defines various types of messages (e.g., GetRequest, GetNextRequest, GetBulkRequest, SetRequest, Response, Trap), and their respective Functions.

    SNMP Protocol: Message Formats

    • Explains the formats for different message types (e.g., Get/set header, Trap header, and SNMP PDU).

    SNMP: Management Information Base (MIB)

    • Briefly discusses MIB modules, Structure of Management Information (SMI), and example MIB variables for UDP protocol

    NETCONF Overview

    • Describes the function of NETCONF, which facilitates managing network devices using an XML-based, RPC-based mechanism.

    Selected NETCONF Operations

    • Explains various NETCONF operations used for handling and managing network configurations.

    Sample NETCONF RPC Message

    • Provides example NETCONF RPC message formats.

    YANG Data Modeling Language

    • Explains YANG, a data modeling language that is used in network configuration, and is often used in conjunction with NETCONF.

    Network Layer Summary

    • Summarizes the essential concepts and elements involved in network layer control components.

    Additional Chapter 5 Slides

    • Notes about extra slides in chapter 5.

    Distance Vector: Another Example

    • Shows another distance vector calculation in an example.

    Studying That Suits You

    Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

    Quiz Team

    Related Documents

    Chapter 5 Network Layer Control Plane PDF

    More Like This

    Untitled Quiz
    6 questions

    Untitled Quiz

    AdoredHealing avatar
    AdoredHealing
    Untitled Quiz
    37 questions

    Untitled Quiz

    WellReceivedSquirrel7948 avatar
    WellReceivedSquirrel7948
    Untitled Quiz
    55 questions

    Untitled Quiz

    StatuesquePrimrose avatar
    StatuesquePrimrose
    Untitled Quiz
    50 questions

    Untitled Quiz

    JoyousSulfur avatar
    JoyousSulfur
    Use Quizgecko on...
    Browser
    Open
    Browser
    Browser