Intro Computer Nets
What is Dijkstra's Algorithm?
A shortest path algorithm used to find the shortest distance between two nodes in a network.
What are all the real time protocols?
SDP/SMIL RTSP RTP/RTCP RSVP/DiffServ
What are the main important parts of the TCP header?
Source port, Destintaion port, Sequence Number, Acknowledgment number, Window Size, and Maximum Segment Size
How is a TCP connection established?
Through a 3 way handshake. 1. SYN 2. SYN-ACK 3. ACK
What are the two types of DNS queries? Why Do we break up DNS servers into this heierachy?
* Iterated and Recursive. * We break it up into a heriarchy because it allows for the work to be split up amoung the servers and reduces security risks for things like DNS ddos.
What is the backbone of the internet?
* The backbone consists of networks from a number of interconnected big Network operators * They are called Tier 1 operators
List the TCP/IP layers and explain each.
Application Layer: This is the topmost layer of the TCP/IP model. It is responsible for handling communication between the user and the network. The applications at this layer interact with the network through protocols like HTTP, FTP, SMTP, and others. This layer deals with issues such as data formatting, encryption, compression, and user authentication. Transport Layer: The transport layer is responsible for the end-to-end delivery of data between hosts. It ensures that data is delivered reliably and efficiently. The most common protocols at this layer are TCP and UDP. TCP provides reliable, ordered, and error-checked delivery of data, while UDP provides faster, unreliable delivery. Internet Layer: The internet layer is responsible for the delivery of data between hosts on different networks. It handles issues such as routing, addressing, and packet fragmentation. The most commonly used protocol at this layer is the Internet Protocol (IP). Link Layer: The link layer is responsible for the physical transmission of data over a network. It deals with issues such as framing, error detection, and flow control. The link layer is divided into two sublayers: the Media Access Control (MAC) layer and the Logical Link Control (LLC) layer.
What is the basic architectire of a SDN? and what are some SDN functionalities? What are the two things you absolutely need from the SDN stack?
Control logic comes from the northbound interface and then is sent to SDN routers from the controller through the southbound interface. ** Note you can connect a bunch of controllers to connect more SDN routers So there is Controller to network communication and Controller to Controller communication in a SDN * End to end layer 2 paths * Forwarding based on source as well as destination * All traffic from a specific MAC address sent along a specific path * Segregation of traffic based on application type * Multi-path forwarding based on has of 4-tuple (IP sender/destination and port sender/destination) * Transport of non standard layer 3 protocols * You at least need hardware and the controller from the SDN stack
Explain Base band and pass band
Difference between meaning: Base Band- as the name suggests refers to the original transmission signal generated from the message source with modulation of high frequency carrier whatsoever. Pass Band signal- on the other hand, is yet another form of digital transmission that uses modulation meaning the frequency or phase of the carrier signal is modulated to transmit the bits. It is based on the concept of digital data transmission post modulation of high frequency sinusoidal carrier.
What is the industry standard model for Network management?
FCAPS F- Fault detection and correction C- Configuration and operation A- Accounting and billing P- Performance assesment and optimization S- Security assurance and protection
How does IPv6 work?
Instead of fragmenting packets it just drops packets that are too large and signal packet too large. End nodes are meant to do path MTU discovery to determine maximum packet size.
Why do we need to tear down a TCP connection?
It is needed for both computation devices to agree they ghave no more data to send before they can disconenct. A FIN is used to ask to close connection.
What is congestion and what are some ways to deal with it?
It is the biggest issue in any network and is un-avoidable. It occurs when there is more traffic then the network can handle. Its best effort, so everyone send and whoever gets dropped gets dropped. The ways to handle it are below: 1. Circuit Switching 2. Dropping Packets 3. The server may use reservations 4. Pricing: do not drop packets from the high bidders 5. Dynamic adjustment (TCP): Every sender infers the level of congestion and adpats its sending rate for the greater good.
Explain some common routing issues
Loops * Can be caused if good news flows backwards Slow convergence * Routes fail to converge after a change *Example: A says it can connect to internet and B now can conenct indirectly to the internet throuigh A. A can now no longer connect to the internt But B says it can. Black Hole * Routing system sends packets for a set of destinations to a location where they are silently discarded Route Flapping * Routes continue to oscillate * Can be caused by equal-length paths
Can an EGP report the cost of a path?
No it only can know of its existance. Each ISP is an AS that uses an EGP to advertise its customers networks to other ISPs.
What are 3 reasons for retransmission?
Packet lost, Ack lost, Early Timeout
What is a conselation diagram?
* A constellation diagram is a plot of the phase and relative amplitude of the output symbols for a digital modulation system, in polar coordinates. In terms of the symbols phase 0 degrees is along the positive x-axis, and phase increases as you move counterclockwise around the x-y plane. * The symbol relative amplitude is measured as distance from the origin of the plot. The possible output symbols are represented with large dots, and adjacent to them are the bits they represent. Symbols that have the same amplitude are the same distance from the origin (you can think of them as laying on the same circle around the origin). All symbols with the same phase would fall on the same line segment that originates at the origin and goes out at a certain angle.
What are the two ways to measure network performance and why is it difficult to do in general?
* Active- by injecting traffic into a network, the measurement traffic affect the performance of the network * Passive- monitor a network and count packets. Does not inject * Measuring network performance can be difficult because of the following: * Routes can be asymmetric * Conditions change * Measurement can affect performance * Traffic is bursty
Explain what an AS (Autonomous System) is
* An autonomous system is a set of networks and router under one administrative authority * Flexible, soft definition * Intuition: a single corporation * Needed because no routing protocol can scale to entire Internet * Each AS chooses a routing protocol
List the main protcols in each OSI layer
* Application- FTP, HTTP/S, SMTP, Telnet * Presentation- Data translation/compression, encryption/decryption * Session- Authorization, Session manegment, SCP, SQL, NFS * Transport- TCP, UDP, Flow Control, Segmentation, Error Control, Logical Communication, Multi/de-multiplexing * Network- Logical Addressing (IP), Ruting, Path determentation, ARP, IPv4/IPv6, ICMP * Data Link- Ethernet, Framing * Physical- The medium, Modulation de-modulation (AM, FM, QAM, M-ARY PSK)
List the Specific Routing protocols
* BGP * RIP * OSPF * IS-IS
What are CSMA, CSMA/CD, CSMA/CA?
* Carrier sense multiple access (CSMA) is a media access control (MAC) protocol used in local area networks (LANs) that uses carrier sensing to detect when the network is in use. It is a type of multiple access protocol that allows multiple devices to share a single communication channel. When a device wants to send data, it first listens to the channel to make sure that no other device is currently sending data. If the channel is clear, the device can then send its data. If the channel is busy, the device must wait until it is free before sending its data. * In CSMA if 2 terminals begin sending packet at the same time, each will transmit its complete packet (although collision is taking place). Wasting medium for an entire packet time CSMA/CD Step 1: If the medium is idle, transmit Step 2 If the medium is busy, continue to listen until the channel is idle then transmit If a collision is detected during transmission cease transmitting Wait a random amount of time and repeat the same algorithm * CSMA/CA CSMA with collision avoidance All terminals listen to the same medium as CSMA/CD Terminal ready to transmit senses the medium If medium is busy waits until the end of current transmission It again waits for an additional pre-determined time period DIFS (Distributed inter frame Space). Then picks up a random number of slots (the initial value of back off counter) within a contention window to wait before transmitting its frame If there are transmission by other terminals during this time period (back off time) the terminal freezes it counter It resumes count down after other terminals finish transmission + DIFS. The terminal can start its transmission when the counter reaches zero Inside of CSMA/CA there is RTS-CTS exchange (request to send and clear to send)
What are two ways of handling jitter? and how do we measure it?
* Design an isochronous network with no jitter * Traditional telephone systems used this. Build a network with no jitter * Use a protocol that compensates for jitter * Voice and video over the internet takes this approach. This is much less expensive * The technique for measuring jitter is a sender emits a series of packets with a small fixed delay between packets and the receiver records the time at which each packet arrives and uses the sequence of times to compute the differences in delay
Explain modulation
* Essentially you modify a normal carrier signal with your signal you want to send in order for the carrier signal to carry your data. The carrier frequency should be more than the modulation frequency. * The base band signals are incompatible for direct transmission. For such a signal to travel longer distance its strength has to be increased by modulating with a high frequency carrier wave, which does not affect the parameters of the modulating signal.
Explain fine grained and coarse grained QoS and what is the goal of Traffic scheduling?
* Fine grained is when a connection is made the agreement is just for this communication * Flow generally refers to transport layer communication * Four main categories of service (Constant Bit Rate, Variable Bit Rate, Available Bit Rate, Unspecified Bit Rate) * VBR is hard to meet there are a lot of things to specify. * Sustained Bit Rate (SBR), Peak Bit Rate (PBR), Sustained Burst Size (SBS), Peak Burst Size (PBS) * ABR is what is used for the most part today * Coarse grained we look at type of packet to determine what quality it is agreed to get. (Works way better for the Internet fine grain doesn't work on Internet) * Traffic is divided into classes and it does not need to be in a flow each packet just must be marked. * This is good because routers deal with millions of flows which is complex and resource intensive. * Traffic scheduling using Round robin or priority or etc. The goal is to prioritize some type of traffic while being fair.
What are the main multiple access protocols?
* Frequency Division Multiple Access (FDMA)- Different frequencies communications at same time. * Time Division Multiple Access (same frequency and divide up the time for the separate communications happening) * Code Division Multiple Access multiple signals in the same frequency and in the same time slot. Each signal uses a different code. Originally for military use. * Random Access: In random access or contention methods, no station is superior to another station and none is assigned the control over another. At each instance, a station that has data to send uses a procedure defined by the protocol to make a decision on whether to send.
What are the 4 types of routing?
* Global- all routers have complete topology. Link state algorithms * Static- routes change slowly over time * Dynamic- routes change quicker. Periodic update or in response to link cost changes. * Decentralized- iterative process of comutation, exchange of info woth neighbors. Routers initially only know link costs to attached neighbors. Distance vector algorithms.
What are some routing metrics? And how are Administrative costs assigned?
* It is possible to use throughput, delay, jitter or any other measurable property as a routing metric * But most internet routing software does not uses them as is * Typical internet routing uses a combination of two metrics: Administrative cost and hop count * Administrative costs are assigned manually * Often to control which paths traffic can use * Routing software will choose the path with the lower cost * Thus traffic will follow the corporate policy
What is an Element Management System? and what are some network management tools?
* Management tool that can manage one element at a time * Typically, supplied by vendor of the network element * Unfortunately many networks only have element management Tools Physical Layer Testing Reach ability and Connectivity Packet Analysis Network Discovery (map a network) Device Interrogation Event Monitoring Performance monitoring Flow analysis Routing and Traffic Engineering Configuration Security Enforcement
How do we typically measure delay and why?
* Measuring Delay is hard * It requires that the two hosts have synchronized clocks * Instead of using synchronized clocks many network measurements tool choose an easier approach * Measure the round trip time and divide by two
Explain some key details of IS-IS (not the terrorist organization)
* Originally designed by DEC to be part of DECNET V * The IS-IS is an IGP * Naming follows Digital terminology in which a router was called as IS and a host was called an End system * IS-IS was created around the same time as OSPF * The two protocols are similar in many ways * Both use link-state approach * Employ Dijkstra's algorithm to compute shortest paths * Both protocols require two adjacent routers to periodically test the link between them and broadcast a status message * The main differences between OSPF and IS-IS can be the following: * IS-IS was proprietary * And OSPF was create as an open standard * OSPF was designed to run over IP * IS-IS was designed to run over CLNS * OSPF was designed to propagate IPv4 routes * IS-IS was designed to propagate router for OSI protocols * Over time, OSPF gained many features * As a result IS-IS now has less overhead * When the protocols were initially invented * OSPS's openness made it much more popular * Over all IS-IS can now be seen as a lighter weight but less feature rich alternative to OSPF
Explain BGP and BGP hijacking
* Provides routing among AS's (EGP) * Policies to control routes advertised * Uses TCP * Gives path of autonomous systems for each destination * Currently the EGP of choice in the internet * Supports Classless Inter-Domain routing and uses route aggregation to decrease the size of routing tables BGP, or Border Gateway Protocol, is a routing protocol used to exchange routing information between different networks on the Internet. It is used to connect different autonomous systems (AS) or networks and help them communicate with each other. BGP is a path-vector protocol, which means that it selects the best path for a packet based on a set of rules that take into account the number of hops, bandwidth, and other factors. It allows network administrators to control the flow of traffic in and out of their networks, and to establish policies for routing traffic to specific destinations. BGP Hijacking * If you advertise access to networks through BGP, you can get large parts of traffic to theses networks routed through your network. This makes it possible to monitor and eavesdrop this traffic.
What are the ALOHA multiple access protocols?
* Pure ALOHA- Additive links on line Hawaii area protocol If there is a collision, the sender waits a random amount of time and sends it again. The waiting time must be random otherwise the same packets will collide again. * Slotted ALOHA- Slotted ALOHA divides the time slots equal to packet transmission time. A node transmits at the beginning of the next slot. If collision wait for a future slot and decreases the risk of collision. Synchronous that is time is divided into slots. Slot size is equal to the transmission time of a packet When you are ready, transmit at the start of the time slot Doubles the efficiency of ALOHA (38% throughput) But requires synchronization
What is Forwarding / the data plane?
* Refers to data gram transfer * Performed by host or router * Uses routing table
What is Routing / the control plane?
* Refers to propagation of routing information * Performed by routers * Inserts/changes values in routing table
What are some key reasons intuition can fail you when handling Routing?
* Routing is not like water flowing through pipes * Multi-path routing is difficult * Capacity can go unused if not along shortest path * Fewest hops may not always be best * Routing around congestion is not straightforward, and did not always yield a big improvement * Can cause out of order packets * Can result in route flapping
Explain OSPF
* Routing within an autonomous system (IGP) * Full CIDR and subnet support * Authenticated messages exchanges * Allows routes to be imported from outside the AS * Uses link-status (SPF) algorithm * Support for multicast networks e.g. Ethernet * OSPF Area and Efficiency * Allows subdivision of AS into areas * Link-status information propagated within area * Routes summarized before being propagated to another area * Reduces overhead (less broadcast traffic) * Because it allows a manager to partition the routers and networks in an AS into multiple areas, OSPF can scale to handle much larger number of routers than other IGPs * Some networks are to big even for OSPF. For these networks BGP can be used as IGP. OSPF stands for Open Shortest Path First, which is a dynamic routing protocol used in computer networking. Like other routing protocols, OSPF helps to determine the best path for sending data between different devices or networks, but it uses a more advanced algorithm than simpler protocols like RIP. OSPF is designed to work well in large and complex networks, and it uses a link-state algorithm to calculate the best path to a destination network based on a variety of factors like bandwidth, delay, and cost. Each router maintains a database of the network topology, which includes information about all of the routers and links in the network, and uses this information to calculate the best path to each destination. Unlike distance-vector protocols like RIP, OSPF exchanges link-state advertisements (LSAs) between routers to build and maintain the network topology database. Each router floods its own LSA to all other routers in the network, and uses this information to build a complete view of the network topology. Based on this information, OSPF calculates the shortest path tree to each destination network, which determines the best path for sending data.
What is sampling?
* Sampling is converting analog media streams to digital * Sample rate refers to the rate at which real-time info has been converted (e.g. 8000 times per second) * Important concept- Because each source of real-time info can choose a sample rate, playback and synchronization must know the sample rate and encoding that was selected.
What are the two ways of Telephone Number Mapping and Routing by the IETF?
* TRIP relies on location servers to exchange information * It is built similar as BGP, but with the purpose to route telephone calls on internet rather than data grams. * In reality, TRIP is not used anywhere. Telephony routing in the different inter-operator peering solutions is achieved using ENUM. * ENUM (E.164 NUMbers) uses arpa top-level domain in the Domain name system. * ENUM (E.164 Number to URI Mapping) translates telephone numbers into Internet addresses. You can dial a telephone number and reach a SIP. H.323 or any other internet telephony user. This all happens in the background; you do not need to do anything special while calling someone. * A server with ENUM support will look up a dialed telephone number in the ENUM tree of the DNS to see if there are alternate ways to set up the call instead of just calling out in the PSTN telephone line. ENUM may contain a reference to SIP URI, a telephone number to dial, a web page or an e-mail address.
How does static routing work?
* Table initialized when system boots * No further changes * Used by most internet hosts * Typical routing table has two entries * Local network —> direct delivery * Default —> nearest router
How does dynamic routing work?
* Table initialized when system boots but the routing software learns routes and updates the table continuously. * Used by IP routers * Requires special software * Each router communicates with neighbors * Pass routing information * Typically use the Link-stat or distance vector algorithm * Use route propagation protocol * Each router advertises destinations that lie beyond it
Explain SDP
* The session description Protocol is a format for describing multimedia communication sessions for the purposes of session announcement and session invitation * Synchronized multimedia integration language is an extensible markup language (XML) markup language to describe multimedia presentations * It defines markup for timing layout animations visual transition and media embedding among other things. SMIL allows presenting media items such as text, images, video, audio, links to other SMIL presentations, and files from multiple web servers.
What are some ways that a network managment system could work?
* Use a parallel physical Network (too costly) * Use a parallel logical network (same wires but separate traffic) * Use a special link-layer protocol * Use the same links, equipment, and protocols as data (most used today)
What are the most important parts of the IPv4 header?
* Version * Internet header length- the size of this field is 4 bits so you multiply the value of this field by 4 to get the total bits. * TOS (Type of service)- It explains using bits how the packet should be handled. * Total Length- length of header + length of data * MTU- the largest size of the IP packet that can be transmitted over a particular network link without fragmentation
Explain IP multi cast and Ethenet delivery plus the issue
* When sending IP multicast across Ethernet * Can use Ethernet multicast capability * IP multicast address is mapped to an Ethernet multicast address * Problem * Most interface hardware limits the number of Ethernet multicast addresses that can be used simultaneously * Trick: use a few multicast addresses and allow software to decide how a given packet should be processed (multiplexing)
What are the 3 approaches for multi cast data gram forwarding?
1. Flood and Prune: This is a technique used by multicast routers to discover and maintain multicast group membership. When a multicast packet is sent to a network, the multicast router initially floods the packet to all connected network segments. If a network segment has no interested receivers, the router "prunes" that segment, and it is no longer included in subsequent multicast transmissions. This technique helps to reduce unnecessary network traffic and improve overall efficiency. 2. Configuration and Tunneling: In this technique, multicast routers are manually configured with information about the multicast group and the location of the multicast source. The multicast traffic is then "tunneled" through the network, and multicast packets are encapsulated within unicast packets to ensure that they reach their intended recipients. This technique is useful in situations where the multicast group is located in a remote network or where multicast routing is not supported. 3. Core Based Discovery: This technique involves the use of a centralized multicast router, called the "rendezvous point," to manage and distribute multicast traffic. When a multicast source sends a packet, it is first sent to the rendezvous point, which then distributes the packet to all interested receivers. This technique is useful in large-scale multicast networks and helps to reduce network traffic by limiting the number of multicast routing updates. *****No mechanism has yet been demonstrated that would allow the IP multicast model to scale to the full Internet. For this reason and also reasons of economics IP multicast is not in general used in commercial Internet backbones.
What are the components of delay?
1. Propagation Delay * A signal require a small amount of time to travel across a transmission medium * Propagation delays are proportional to the distance spanned 2. Access Delay * Many networks use shared media * Set of computers that share a medium must contend for access * Such delays are known as access delays * Access delays depend on * The number of stations that contend for access and * The amount of traffic each station sends 3. Switching delay * The time required to forward a packet 4. Queuing Delay * The time a packet spends in memory waiting to be serviced 5. Server Delay * Not really networking delay but is the processing time needed to service a request
What is Pure ACK?
A "pure ACK" packet refers to an ACK packet that does not contain any data. It is simply a packet with a header that acknowledges receipt of a previous packet, without containing any new data to be transmitted. Pure ACKs are used in situations where the receiver has no data to send back to the sender, but needs to acknowledge receipt of a packet to ensure that the sender knows the data was successfully received. They are typically sent immediately after the receiver receives a packet, without waiting for additional data to send back in the ACK packet.
How do parity and two dimensional parity checks work in error detection?
A parity bit is added to every data unit so that the total number of 1s(including the parity bit) becomes even for even-parity check or odd for odd-parity check
What is ARP and explain it?
ARP or address resoultion protocol is used to reolve IP addresses to MAC addresses. This information is stored in an ARP table.
What type of signals does the physical layer carry?
Analog signals
What are all the layers of the OSI model?
Application, Presentation, Session, Transport, Network, Data Link, and Physical. (Please do not throw sausage pizza away)
What is bandwidth in terms of a analog wave?
Bandwidth is the difference between the highest and the lowest frequencies that the medium can satisfactorily pass.
What are the main difference between CSMA/CD and CSMA/CA?
CSMA/CD- Wired networks, higher efficiency and more popular CSMA/CA- Wireless networks
What is channelization?
Channelization is a multiple-access method in which the available bandwidth of a link is shared in time, frequency, or through code, between different stations
How does checksum error detection work?
Checksum error detection is a technique used to detect errors in data transmission or storage. It involves adding an extra piece of data, called a checksum, to a message. The checksum is calculated based on the content of the message and is designed in such a way that any changes made to the message during transmission or storage will result in a different checksum value. Here is an example of how checksum error detection works: Suppose you have a message that you want to transmit, which consists of a sequence of bytes. To calculate the checksum, you add up all the bytes in the message and take the result modulo some number, typically a power of 2. The resulting value is the checksum. For example, suppose the message is: 01101001 01101110 01110100 01110010 01101111 To calculate the checksum, you add up all the bytes: 01101001 + 01101110 + 01110100 + 01110010 + 01101111 = 100101000 Then you take the result modulo 256 (which is 2^8, since we are dealing with bytes), giving you a value of: 100101000 % 256 = 168 So the checksum for this message is 168. When the message is transmitted, the checksum is also sent along with it. The receiver then calculates the checksum for the received message and compares it to the checksum that was sent. If the two checksums match, the message is assumed to be error-free. If they don't match, then an error has occurred during transmission or storage, and the message must be retransmitted or corrected in some other way.
What is class based addressing and what are the differnt classes?
Class-based addressing is a method of IP address allocation that categorizes IP addresses into different classes based on their network requirements. This approach was used in the earlier days of the Internet, but has since been replaced by Classless Inter-Domain Routing (CIDR), which is a more flexible and efficient system. There are five classes of IP addresses under class-based addressing: 1. Class A: This class of addresses uses the first octet to represent the network portion of the address and the remaining three octets to represent the host portion. Class A addresses start with a binary value of 0 in the first octet and have a range from 1.0.0.0 to 126.0.0.0. 2. Class B: This class of addresses uses the first two octets to represent the network portion of the address and the remaining two octets to represent the host portion. Class B addresses start with a binary value of 10 in the first octet and have a range from 128.0.0.0 to 191.255.0.0. 3. Class C: This class of addresses uses the first three octets to represent the network portion of the address and the remaining octet to represent the host portion. Class C addresses start with a binary value of 110 in the first octet and have a range from 192.0.0.0 to 223.255.255.0. 4. Class D: This class of addresses is reserved for multicast addresses and has a range from 224.0.0.0 to 239.255.255.255. 5. Class E: This class of addresses is reserved for experimental purposes and has a range from 240.0.0.0 to 255.255.255.255. Each class of addresses has a default subnet mask associated with it, which defines the network and host portions of the address. Class-based addressing is no longer used in modern networks, as CIDR allows for more efficient use of IP addresses by allowing variable-length subnet masks.
What is Classless Inter-Domain Routing (CIDR)?
Classless Inter-Domain Routing (CIDR) is a method of IP address allocation and route aggregation that is used to conserve IP address space and reduce the size of routing tables in the Internet. Unlike class-based addressing, which divides IP addresses into fixed-sized classes, CIDR allows for variable-length subnet masks, which means that IP addresses can be allocated and assigned more efficiently. CIDR notation is used to represent IP addresses with variable-length subnet masks. It consists of the IP address followed by a forward slash (/) and a number, which represents the length of the subnet mask. For example, the CIDR notation 192.168.1.0/24 represents a network with an IP address of 192.168.1.0 and a subnet mask of 255.255.255.0. CIDR allows for the allocation of IP addresses in smaller blocks, which reduces waste and allows organizations to more efficiently manage their IP address space. It also enables route aggregation, which means that multiple IP addresses can be represented as a single entry in a routing table. This reduces the size of routing tables and improves the efficiency of routing in the Internet. CIDR is widely used in modern networks and is the standard method of IP address allocation and routing in the Internet today. It is also used in conjunction with other Internet protocols such as Border Gateway Protocol (BGP) to enable efficient routing between autonomous systems (ASes) in the Internet.
What is the point of routing exchange?
Each router runs routing software that learns about destinations other router can reach and informs other router about destinations it can reach. The routing software uses incoming information to update the local routing table continuously.
What is the main concept behind error detection?
Error detection uses the concept of redundancy which means adding extra bits for detecting errors at the destination. To be able to detect or correct errors, we need to send some extra bits with our data. These redundant bits are added by the sender and removed by the receiver. Their presence allows the receiver to detect or correct corrupted bits. The concept of including extra information in the transmission for error detection is a good one. But instead of repeating the entire data stream, a shorter group of bits may be appended to the end of each unit. This technique is called redundancy because the extra bits are redundant to the information: they are discarded as soon as the accuracy of the transmission has been determined.
What has a higher bandwith AM of FM?
FM uses a higher frequency range and a bigger bandwidth that AM... Each FM station is allocated 150 kHz of bandwidth, which is approx 15 times the of an AM station. This means that an FM station can transmit 15 times as much information as an AM station and explains why music sounds much better on FM.
Explain FTP
FTP is file transfer protocol. A main idea of FTP is how it works. It works by each time a new request is made a new conenction is formed between the client and server. For example a connection is formed by the client to request a directory from teh server then this connection closes. A new connection is then formed by the server to send the directory listing to the client.
What layer are routers in, gateways, bridges, and repeaters?
Gateway- Can be in all layers but mainly in and above Transport Router- In Network, data link, and physical Bridge- Data link and physical Repeater- Physical
Whats the main idea between the IETF approach and the ITU approach?
IETF- * Known as Session initiation Protocol (SIP) * Domain name system used to map a telephone number to an IP address ITU- * Standard is H.323 * Differs substantially from terminology used by SIP * Terminal provides IP telephone functions and may also include facilities for video and data transmission * Gatekeeper provides location and signaling functions, and establishes connections to the public switched telephone network (PSTN) * Gateway interconnects the IP phone system and PSTN, and handles both signaling and media translation * Multipoint Control Unit (MCU) provides services such as multipoint conferencing
Explain IPv4 multicast and IPv6 multicast
IPv4- In IPv4, multicast addresses are defined within the Class D range of IP addresses, which includes addresses from 224.0.0.0 to 239.255.255.255. These addresses are reserved for multicast use and cannot be assigned to individual devices. When a device sends a multicast packet, it sends it to a specific multicast address rather than a unicast address. Multicast communication uses a protocol called Internet Group Management Protocol (IGMP) to manage membership in multicast groups. When a device wants to receive multicast traffic, it sends an IGMP message to join a multicast group. The sender then sends packets to the group address, and all devices that have joined the group will receive the packets. IPv6 - IPv6 prohibits broadcast, but defines multicast groups that are equivalent
What is selective repeat ARQ?
In Selective Repeat ARQ, the sender divides the data into a sequence of packets and sends them to the receiver. Each packet is numbered, and the receiver sends an acknowledgment (ACK) for each correctly received packet. If a packet is lost or corrupted during transmission, the receiver sends a negative acknowledgment (NACK) to the sender, indicating that the packet needs to be retransmitted. Unlike other ARQ protocols like Go-Back-N ARQ, Selective Repeat ARQ allows the sender to continue sending packets while waiting for NACKs for the lost or corrupted packets. The sender only retransmits the lost or corrupted packets, rather than retransmitting all the packets in the sequence. This reduces the amount of retransmitted data and increases the efficiency of the transmission.
Explain what a spanning tree is
In computer networking, a spanning tree is a type of network topology that is used to prevent loops in a network. It is a subset of a network that includes all the nodes of the original network, but with only enough links to ensure that there are no loops.
What are the three controled access algorithms?
In controlled access the stations consult one another to find which station has the right to send. A station cannot send unless it has been authorized by other stations. * Token passing- is a broadcast-based technology because all stations see every frame. The forwarding of both the token and data frames is performed by NICSs in hardware. A token circulates among all stations. The token is a miniature, 3-byte frame (including start and end flags). When the token arrives a station either seizes the token and sends a frame or passes the token to the next station. Assuming that station i has a frame f to send to station j. 1. Station i waits for the arrival of the token and seizes the token 2. Station i sends f to station i+1 which in turn passes f to station i+2 3. When frame f arrives at station j, station j picks up f and simultaneously forwards f to station j+1 4. Eventually f returns to station i which passes the token rather than f to station i+1 * Whenever the primary station wants to receive the data it asks the secondary stations present in its channel this method is polling. In the first diagram we see that primary station asks station A if it has any data ready for transmission since A does not it send NAK until it is polled again and does have data to send queued up. * Reservation: Whenever we travel from a train or an airplane, the first thing we do is to reserve our seats, similarly here a station must make a reservation first before transmitting and data-frames This reservation time line consists of two kinds of period: 1. Reservation interval of a fixed time duration 2. Data transmission period of variable frames
What is piggybacking?
In piggybacking, instead of sending a separate acknowledgment (ACK) packet for every packet received, the receiver of the packets waits for a small amount of time to see if there is any data to send back to the sender. If there is, the ACK packet is sent along with the data, in the same packet, effectively "piggybacking" on the sender's packet.
What is Open Flow?
It is the SDN standard now. OpenFlow works by defining a set of standard messages that the controller and the switches use to communicate. The controller can send messages to the switches to modify their forwarding behavior based on network conditions or policies. The switches then forward packets based on the rules programmed by the controller. * Examines the packet for multiple patterns at a time and what ever pattern is matched first its action is executed. This is very fast and can happen at the speed of any network. * If no patterns match it needs to report to the controller it has a new type of packet it does not know how to handle. So the controller can make new rules for it so it does not have to drop that packet.
How do you calculate the number of redundancy bits (r) needed to correct a given number of data (m)?
It must be able to represent m + r + 1 different states
How long should TCP wait to retransmit?
It should set the timeout based off the RTT (round trip time) and a fudge factor. The two relevant equation are below. * TimeoutInterval = EstimatedRTT + 4*DevRTT * DevRTT = (1-Beta)*RecentlyDevRTT + Beta* |SampleRTT-Estimated RTT|
Explain M-ary PSK
M-ary PSK (Phase Shift Keying) is a digital modulation technique that extends the basic PSK technique to represent multiple bits per symbol. The M in M-ary PSK refers to the number of distinct phases that can be used to represent digital data. In M-ary PSK, the carrier signal is modulated by shifting the phase of the signal to one of M different values, where M is typically an even integer (e.g. 2, 4, 8, 16, etc.). Each of the M phase shifts represents a unique combination of bits. For example, in 4-ary PSK, each phase shift can represent one of four possible bit combinations: 00, 01, 10, or 11.
What is MIB?
MIB stands for Management Information Base. In network management, a MIB is a database of objects that can be managed by a network management protocol, such as SNMP (Simple Network Management Protocol). The MIB contains information about various aspects of a network device, such as its configuration settings, operational status, and performance metrics. Each object in the MIB is identified by a unique object identifier (OID), and has a corresponding value that can be queried, set or monitored by a network management system. * Are hierarchical * Begin with standard prefix *** Note all MIB variables start with this iso.org.dod.internet.mgmt.mib * Name is encoded as integers: Use GET-NEXT operation to move through array of variables
What is a burst bit error?
Means that 2 or more consecutive bits in the data unit have changed
What is NAT and what does it do?
NAT or network address translation is used to connect computers on local provate networks across the internet. It goes from both private to public and visa versa. If there was only one computer on the local network NAT would not be needed.
Explain QPSK (Quadrature Phase Shift Keying)
QPSK (Quadrature Phase Shift Keying) is a type of Phase Shift Keying (PSK) modulation technique used in digital communication systems. QPSK is a form of PSK that uses four different phases to represent two bits of digital data. Each phase shift represents a specific combination of two bits (00, 01, 10, or 11). In QPSK, the carrier signal is split into two components, referred to as the I (in-phase) and Q (quadrature) channels. These two channels are modulated separately using the four-phase shift, and then combined to create the QPSK signal. QPSK provides higher data rates than standard binary PSK modulation techniques because it transmits two bits of information per symbol. However, QPSK is also more susceptible to errors due to noise and interference because it is transmitting two bits of information per symbol, and any error in the received signal will affect both bits.
Explain QAM
Quadrature amplitude Modulation (QAM): *In order to increase the distance between symbols in a constellation, another option is to modulate both the amplitude and the phase of the carrier.
Explain RIP
RIP stands for Routing Information Protocol, which is a dynamic routing protocol used in computer networking. The main function of RIP is to allow routers within a network to share information about their available routes, which helps to determine the best path for sending data between different devices or networks. In RIP, routers exchange information about their routing tables with their neighboring routers using a standard message format. These messages include information about the destination network, the metric or cost of the path to reach that network, and the next hop router that should be used to forward data to the destination. RIP uses a distance-vector algorithm to calculate the best path to a destination network, based on the number of hops or routers that need to be traversed. Each router maintains a routing table that lists all of the available networks and their associated metrics, and periodically updates this table based on the RIP messages it receives from neighboring routers. One of the key limitations of RIP is that it has a maximum hop count of 15, which means that it may not be suitable for large or complex networks. Additionally, RIP does not support some advanced features like load balancing or traffic engineering, which are available in more advanced routing protocols like OSPF and BGP.
Explain the following Internet QoS technologies RSVP/COPS/IntServ, DiffServ, and MPLS.
RSVP/COPS/IntServ * Fined grained * Reserve resources along the path. All devices along the way need to support this. COPS handles error reporting and policing for RSVP. DiffServ * Coarse-grained * Uses Type of Service field for prioritization. This is not widely accepted so there is no guarantee the priority will be acknowledged. MPLS (Used in tier 1 ISPs) * At one end of the path a data gram is encapsulated in an MPLS header and injected into the MPLS path * At the other end each data gram is extracted, the MPLS header is removed and the data gram is forwarded to its destination * It is a connection-oriented mechanism that is a layer 2.5 protocol. It does not break anything. The ISP might set up an MPLS path for voice data that is separate from the MPLS path used for other data. Uses MPLS routers.
Explain RTP and what protocol is used like a tv remote to control a RTP stream?
Real time transport Protocol * Allows sender and receiver to choose sample rate and encoding * Specifies a header for each message transferred * Uses UDP for transport * Separates time stamp from packet sequence number * Includes a marker bit that allows some frames to be marked * RTP has a companion protocol known as RTP Control Protocol (RTPCP) which provides quality of service feedback to the participants of an RTP session. Marking * Permits differential encoding with a full frame followed by incremental changes. Instead of sending full frame send the difference. When a new frame is sent it can be marked as such. * Real Time Streaming Protocol (RTSP) used to control a RTP stream like a TV remote
What is shift keying and what are the three main ways of shift keying?
Shift keying is a digital modulation technique used in telecommunications that involves modulating a digital signal by shifting its frequency, phase or amplitude to represent binary data. The technique is commonly used in radio communication systems and other digital communication systems to transmit data over long distances. There are three main types of shift keying: 1. Amplitude Shift Keying (ASK): In ASK, the amplitude of the carrier signal is varied to represent binary data. A high amplitude represents a binary 1, while a low amplitude represents a binary 0. 2. Frequency Shift Keying (FSK): In FSK, the frequency of the carrier signal is varied to represent binary data. A high frequency represents a binary 1, while a low frequency represents a binary 0. 3. Phase Shift Keying (PSK): In PSK, the phase of the carrier signal is varied to represent binary data. A change of 180 degrees in phase represents a binary 1, while no change in phase represents a binary 0. Shift keying is a simple and efficient way to transmit digital signals over a communication channel. However, it is susceptible to noise and interference, which can cause errors in the received signal. To reduce errors, error-correcting codes and other techniques are used in conjunction with shift keying.
Explain SNMP
Simple Network Management Protocol (SNMP) * Internet Standard * Allows software in a managers computer (manager)to interact with software that runs in an element (agent) * Specifies format and meaning of messages exchanged * Runs as an application protocol over TCP or UDP * Uses fetch-store paradigm * Every element you manage is essentially a set of variables. * Set of conceptual variables defined * Each variable given a name * Set of variables known as management Information Base (MIB) * SNMP offers two basic operations * GET to read the value of a variable * PUT to store a value into a variable * All management functions are defined as side-effects of GET or PUT to a MIB variable * Example: reboot defined as side-effect of PUT SNMP encoding shown in image
Explain Time To Live
Specifies how long the data-gram is allowed to live on the network, in terms of router hops. Each router decrements the value of The TTL field.
What do ITU and IETF agree and disagree on in IP telephony?
Standards agree on two basics * Audio encoded using Pulse code modulation (PCM) * RTP used to transfer digitized audio0 Standards disagree on * Signaling (Telecommunication term for the process of establishing and terminating a call) * Public Switched Telephone Network (PSTN) interaction
What is TCAM
TCAM stands for Ternary Content Addressable Memory, and it is a type of high-speed memory that is commonly used in SDN routers to perform fast packet lookups and forwarding decisions. In SDN routers, the TCAM is typically used to store the flow tables, which contain the rules that define how incoming packets should be processed and forwarded. The TCAM allows the router to quickly search the flow table and identify the matching rule for each incoming packet, which determines how the packet should be forwarded. The advantage of using TCAM in SDN routers is its fast lookup speed, as TCAM can search the flow table in parallel and identify the matching rule in a single clock cycle. This is much faster than performing a sequential search of the flow table using traditional memory. However, TCAM is also relatively expensive and power-hungry compared to other memory types, which can limit its scalability and efficiency in large-scale SDN networks. As a result, some SDN router vendors may use a combination of TCAM and other memory types, such as SRAM or DRAM, to optimize the performance and cost of their routers.
What are the important parts of TCP flow control?
The MSS and the window size. In TCP flow control, the Maximum Segment Size (MSS) and Window Size are important parameters that help regulate the amount of data that can be sent between two communicating hosts. The MSS is the maximum amount of data that can be carried in a single TCP segment. It is negotiated between the two hosts during the TCP three-way handshake and is typically determined by the maximum transmission unit (MTU) of the network path between them. The Window Size, on the other hand, is the amount of data that the receiver can buffer before sending an acknowledgement to the sender. It is also negotiated during the TCP three-way handshake and can be dynamically adjusted during the session based on the current network conditions. The MSS and Window Size work together to ensure that the sender does not overwhelm the receiver with too much data. If the sender sends data faster than the receiver can handle it, the receiver will start dropping packets, which can lead to degraded network performance. By limiting the amount of data that can be sent in a single segment (MSS) and regulating the flow of data with the Window Size, TCP flow control ensures that data is transmitted at a rate that is sustainable for both the sender and receiver, optimizing network performance and avoiding congestion.
Explain a siding window
The concept behind a sliding window is to allow a sender to transmit a certain amount of data before waiting for an acknowledgment (ACK) from the receiver. The sender divides the data into small packets and sends them one by one to the receiver. The receiver then sends back an acknowledgment for each packet received successfully. The sliding window technique allows the sender to keep transmitting packets up to a certain limit, even if some packets are still in transit or have not yet been acknowledged by the receiver. The sliding window size is determined by the amount of data that can be transmitted without causing network congestion or buffer overflow. As the data packets are sent and acknowledged, the window "slides" forward to allow for the transmission of additional packets. This technique helps to optimize network performance and reduce the amount of time spent waiting for acknowledgments, resulting in faster data transmission and better network utilization. It is widely used in protocols such as Transmission Control Protocol (TCP) to ensure reliable data transmission over the internet.
How do you decide network address part and host address part of an IP address?
The decision of the network address part and host address part of an IP address is based on the subnet mask associated with the IP address. The subnet mask is a 32-bit value that is used to divide an IP address into a network address and a host address. The subnet mask consists of a series of contiguous ones (1s) followed by a series of contiguous zeros (0s). The ones define the network portion of the address, while the zeros define the host portion of the address. For example, a subnet mask of 255.255.255.0 has 24 contiguous ones followed by 8 contiguous zeros. This means that the first 24 bits of the IP address represent the network portion of the address, while the last 8 bits represent the host portion. To determine the network address and host address of an IP address, you perform a bitwise AND operation between the IP address and the subnet mask. The result of this operation is the network address. For example, if you have an IP address of 192.168.1.100 and a subnet mask of 255.255.255.0, you would perform the following bitwise AND operation: IP address: 192.168.1.100 = 11000000.10101000.00000001.01100100 Subnet mask: 255.255.255.0 = 11111111.11111111.11111111.00000000 Result: 192.168.1.0 = 11000000.10101000.00000001.00000000 The result of this operation is the network address, which is 192.168.1.0. The remaining bits of the IP address represent the host address, which in this case is 0.0.0.100. In summary, the decision of the network address part and host address part of an IP address is based on the subnet mask associated with the IP address. The ones in the subnet mask define the network portion of the address, while the zeros define the host portion.
What is Window size?
The information receivng capactity. It is the amount that can be sent without an ACK, because the receiver can buffer. The reciever advertises window to clioent. It tells the amount of free space left, and the sender agrees to not exceed this amount.
What is Stop and Wait ARQ? and what is ARQ?
The simplest way to implement ARQ, the sender sends a packet and waits for its acknowledgement from the receiver. The trick is to determine the timeout, too small and you have too many retransmissions, too large, and you add delays. Has a low performance. ARQ stands for Automatic Repeat reQuest, which is a protocol used in networking to ensure the reliable delivery of data packets. ARQ involves the sender of the data packet receiving feedback from the receiver regarding whether the packet was received correctly or not. If the packet was received correctly, the sender sends the next packet. If the packet was not received correctly, the sender retransmits the packet until it is received correctly.
What is go back and ARQ?
The window size must be less that 2^m because it may accept the wrong frame because it can be numbered the same. When one frame is lost the receiver will block and wait to receive the lost frame.
What are the LAN Topologies, and explain each
There are several different LAN (Local Area Network) topologies, each with its own advantages and disadvantages. The four most common LAN topologies are: 1. Bus Topology: In a bus topology, all devices are connected to a single cable or bus. Each device communicates directly with every other device on the bus. Data is transmitted in both directions along the bus, and devices listen for their specific address before accepting the data. This topology is easy to install and inexpensive, but it can be prone to data collisions and can be difficult to troubleshoot if there is a problem. 2. Star Topology: In a star topology, each device is connected to a central hub or switch using a separate cable. Data is transmitted from the sender to the hub, which then forwards it to the intended recipient. This topology is easy to install and manage, and it provides good performance and scalability. However, it can be expensive, as it requires more cabling and equipment than other topologies. 3. Ring Topology: In a ring topology, each device is connected to two neighboring devices, forming a ring. Data is transmitted in one direction around the ring, with each device forwarding the data to the next device until it reaches its destination. This topology is reliable and provides good performance, but it can be expensive and difficult to install and manage. 4. Mesh Topology: In a mesh topology, every device is connected to every other device in a full or partial mesh configuration. Data is transmitted directly between devices using the shortest path. This topology provides excellent performance, reliability, and fault tolerance, but it can be very expensive and difficult to manage, as every device must be connected to every other device. In addition to these topologies, there are also hybrid topologies that combine elements of two or more topologies. For example, a tree topology combines a star topology with a bus topology, while a partial mesh topology combines elements of a mesh topology with a star or bus topology. The choice of LAN topology depends on the specific needs of the organization, including factors such as cost, performance, scalability, reliability, and manageability.
What are some of the specific throughput measurments possible? How do we measure it?
Throughput- a measure of the rate at which data can be sent through the network (Bits per second). Throughput is a measure of capacity not speed. Imagine a network to be a road between two locations and packets traveling across the network to be cars traveling down the road Throughput determines how many cars can enter the road each second And the propagation delay determines how long it takes a single car to travel the road from one town to another * Capacity of a single channel * Aggregate capacity of all channels * Theoretical capacity of the underlying hardware * Effective data rate achieved by an application * To asses throughput a sender transfers a large volume of data and a receiver records the time from the start of data arriving until all data is there. It then calculates the throughput as the amount of data sent per unit of time.
What are some useful ways to deal with the issues of transmitting real time data across the internet?
Timestamps * Provided by sender * Assigned to each piece of data * Allow receiver to know when data should be played * Use relative values to avoid need for clock synchronization Jitter buffer * Used by receiver * Accommodates small variance in delay * Used by receiver to assemble incoming real time data * Time stamp on an item determines where item is placed in the playback sequence * General principle: ensure info will be available ion time to play without delay * Trick: to compensate for maximum jitter of d, delay playback for d time units * Result: jitter buffer holds just enough data so playback can proceed uninterrupted * During normal operation, playback can continue for d time units while waiting for delayed packets
Why do we use subnets?
We use sub-netting to break down networks into smaller parts which limits the amount of broadcasting happening which saves resources. With 8 bits for example we can have 2^8 -2 hosts because one bit is reserved for the broadcast and router. If we borrow two bits from the host address for the network address that means we can represent 2^2 or 4 sub-networks and the host can now represent only 2^6 - 2 or 62 hosts.
What is TCP fast retransmit?
in some cases, the sender may receive duplicate ACKs for a packet, indicating that the receiver has received the packet multiple times. In this scenario, TCP fast retransmit can be used to quickly recover from the lost packet without waiting for a timeout. When the sender receives three or more duplicate ACKs for the same packet, it assumes that the packet has been lost and retransmits it immediately, without waiting for a timeout. This helps to reduce the delay in recovering from packet loss and improves the overall performance of the TCP connection. TCP fast retransmit is an important feature of TCP, especially in networks where packet loss is common, such as wireless networks. By quickly recovering from packet loss, TCP can ensure reliable data transmission over the network, even in challenging conditions.
What is simplex?
one way communication x--->y not x<-----y
Whats the router do?
passes information from the modem to the devices in the network. The information transferred by a router can be directed to a specific device by its MAC rather than its IP.
