Networks Flashcards

Question

Fiber optic cable

Answer 1

_ long distances at high speeds _ glass or plastic threads _ carry data digitally _ more durable and more secure than electrical (protected from weather) _ single mode cable - one fiber at higher bandwidth and 50x the distance of multi-mode; runs through ocean between continents; more expensive to use _ multi mode cable - multiple fibers; cheaper; <= 500 m; thick fibers; run wider frequency of light up to a few miles; used within a LAN

Answer 2

_ ST - “straight tip connector”; common with multimode until the mid 2000s; replaced by LC _ SC - “standard connector”; larger than LC _ LC - “lucent connector”; smaller than SC; supports more ports per unit space than SC; most common

Answer 3

_ made of copper _ metal shield to block interference, particularly from nearby metals _ used by cable TV companies

Answer 4

_ coax cable _ connect satellite to TV _ modern, higher bandwidth than RG-59

Answer 5

_ coax cable _ thick wire in middle, with shielding _ go long distances because of thick wire _ doesn’t bend well

Answer 6

_ connects two devices of the same type directly to each other _ computers do this via network interface controllers (NICs) or switches _ transmit of one wire connects to receive of the other _ not much used any more

Answer 7

_ connects device to a wall outlet | _ look like crossover and UTP cables

Answer 8

_ 802.11* _ 802.11i (aka WPA2) published in 2004 to replace WPA _ 802.11-2016 - SAE (simultaneous authentication of equals) to exchange the network key; eliminates the need to tell other people of the passphrase in personal mode _ 802.11w - encryption management frames (for WPA3)

Answer 9

_ the higher the frequency, the higher the bandwidth, and the shorter distance it travels _ 2.4 GHz and 5 GHz bands are unregulated _ many devices (microwave ovens, telephones, etc.) operate in the 2.4 GHz range and produce interference

Answer 10

_ media access control | _ has a data link layer address

Answer 11

_ network layer address

Answer 12

_ Internet Control Message Protocol | _ network layer

Answer 13

_ sends an ICMP echo request to a host and listens for a reply _ reports time to get a response (latency) _ times out if no response, indicating either connectivity issues or firewall issues _ -c n sets the ping count _ can ping using IP or FQDN (fully-qualified domain name) or short name

Answer 14

_ command _ trace the route an IP packet takes to a destination _ displays each hop’s (router’s) IP address and latency _ traceroute on Linux; tracert on Windows _ can help show where a ping is failing _ needs superuser privileges

Answer 15

_ command _ like traceroute, shows path of packet from source to destination _ doesn’t need superuser privileges

Answer 16

_ ipconfig on windows _ ifconfig on Linux/UNIX, also running at boot time _ provides all current TCP/IP network configurations of a computer; how interfaces are setup on a given machine _ provides IP, subnet mask, default gateway for each network adapter _ ipconfig shows even disabled devices, ifconfig only shows enabled devices _ /all option shows MAC address, DHCP status, lease info _ release command on windows releases connections and renews adaptors

Answer 17

_ address resolution protocol _ command displays the IP-to-MAC address mappings for hosts in the ARP cache _ addresses discovered by broadcasting request and noting replies _ can add, remove, modify entries in the ARP cache _ can’t broadcast through a router, which retains the MAC address of the router (huh?)

Answer 18

_ network statistics command _ info about active ports _ useful for capacity management _ netstat -r displays routing info for network adapters

Answer 19

_ shows limited DNS info | _ displays name-to-IP-address mappings

Answer 20

``` _ domain information groper (command) _ queries DNS name servers _ receives full DNS response info _ troubleshoots DNS problems _ replacement for nslookup ```

Answer 21

_ shows who owns a domain or a block of IP addresses

Answer 22

_ command _ shows current route tables on a local host _ used by local host to determine where to send traffic _ 0.0.0.0 is the default gateway, where the router sends thing not otherwise defined in the routing table _ add or remove routes

Answer 23

_ secure copy protocol command | _ uses SSH to copy files

Answer 24

_ FTP is unencrypted _ FTPs is encrypted using SSL/TLS (also used by HTTPS) _ uses TCP for reliability _ needed over distances where transmission errors may occur

Answer 25

_ secure sockets layer _ transport layer security _ SSL has an inherent vulnerability, so replaced by TLS _ both exchange a symmetric key via an asymmetric key cipher; the latter being computationally expensive, while the former makes for faster encryption/decryption

Answer 26

_ trivial file transfer protocol _ transfers file from client to server or server to client using UDP, which does not retransmit _ useful on reliable, local networks

Answer 27

_ display info about one or more users on a remote system | _ shows username and last login

Answer 28

``` _ network mapper command _ scans network for hosts and open ports _ used to see what is deployed _ helpful for vulnerability analysis _ not native to any OS, but downloadable ```

Answer 29

_ displays TCP/IP packets and other packers on the network _ a form of protocol analyzer (aka packet sniffer) _ shows packets in human-readable form for troubleshooting or security analysis _ not native to any OS, but downloadable

Answer 30

_ SSH = secure shell; encrypted _ telnet = unencrypted _ manage accounts and devices remotely

Answer 31

``` _ personal area network _ within 20-30 m _ around a person and their devices _ often via Bluetooth or USB _ can act as a gateway to the Internet _ e.g. Apple products ```

Answer 32

_ local area network _ popularized in the 1960s in research labs _ supported TCP/IP in the 1980s _ can be wireless via Wi-Fi _ uses router or switch to manage traffic

Answer 33

_ wireless LAN | _ can act as a gateway to the internet

Answer 34

_ storage area network _ gives servers access to storage devices _ typically dedicated to storage devices _ reduces interference with regular LAN traffic _ might use non-traditional protocols

Answer 35

_ campus area network _ networks multiple LANS over a limited area _ each LAN usually limited to a single building or part of a building _ like a WAN, but smaller geographic area _ one entity usually owns all the LANs _ routers connect each LAN

Answer 36

_ metropolitan area network _ intermediate in size between a CAN and a WAN _ e.g. a city or metro area _ connects LANs owned by different organizations _ usually higher speed than would be available via WAN _ usually connected to a WAN

Answer 37

_ wide area network _ e.g. the internet _ technically, a network spanning political boundaries _ actually, a network that connects smaller networks _ organizations may use private WANs to minimize risk (e.g. banks)

Answer 38

_ aka P2P _ every machine can act as client or server _ e.g. Bitcoin, Tor

Answer 39

_ a line of devices connected by a shared cable _ no longer common _ e.g. connection of a cable modem to a router or a TV to a dish _ ends of the cable must be terminated with special resistors; absent them, the signals bounce back along the wire causing signal reflection and loss of communication _ the network is divided into segments that can be extended _ half-duplex _ all computers see all traffic; the more devices, the more collisions, making it exponentially harder to communicate _ when collisions occur, all devices must wait for the line to clear _ a break in the cable renders the whole network unusable

Answer 40

_ send or receive, but not both at once | _ only one device at a time

Answer 41

_ a token passes around the ring indicating who can use the cable next _ the device with the token does some quick communication and can continue that communication the next time it receives the token _ rather than a single cable, one cable between devices, as the device forwards to the next cable _ more reliable than bus topology, but like bus topology, any break in the cable breaks the network

Answer 42

_ more reliable than single ring _ each computer has two network cards for propagating each ring _ if both rings are broken, the break can readily be fixed by turning the whole network into a single ring by rearranging two unbroken cables _ common in fiber optic networks, such as SONET (synchronous optical network)

Answer 43

_ aka hub-and-spoke network _ a central device (e.g. an Ethernet switch) separately connects to each device on the network _ most common LAN configuration _ cables can be run through walls because they don’t need to go from computer to computer _ break in one cable only affects one device _ central switch is a single point of failure _ some WANs use star topology to connect remote offices to a central location

Answer 44

_ full mesh = each node is connected to all other nodes _ partial mesh = each node connects to a subset of other nodes _ end points in mesh typically use Wi-Fi or radio signals so there aren’t a lot of wires _ routers can be wired in mesh to optimize traffic _ used for high availability _ if one node fails, connections are easily rerouted (aka “self-healing network”) _ e.g. the internet; wireless networks at home

Answer 45

_ can make sophisticated apps available on low-memory, low-clock-cycle devices _ can scale processing power as a function of need, not limited by the power of the end point _ can more easily secure all the data _ can more easily back up all the data _ data produced by one node need not be compatible with another node

Answer 46

_ can operate without a network connection _ no single point of failure affecting everyone _ (not listed, but can’t be controlled by a few people)

Answer 47

_ readily install new (virtual) devices without needing more hardware _ move (virtual) devices without physically moving them _ increase app reliability by eliminating its dependence on hardware _ create whole app point-in-time snapshots for backup and recovery purposes _ change the computing, storage, and networking resources of the device dynamically as needed _ improved hardware utilization (previously, servers usually operated at 5%-15% utilization)

Answer 48

_ software that enables virtual hardware/machines _ can be, but need not be, an OS in itself (the “host” OS) _ the software emulates the hardware _ “guest” OSs run on the virtual hardware _ e.g. VMware, Microsoft Hyper-V, Linux KVM, Citrix XenServer

Answer 49

_ aka “bare-metal” hypervisor _ an OS that dedicates the machine to virtualization _ very little UI; administered from another computer via a web-based portal _ e.g. VMware ESXi, open source KVM

Answer 50

_ aka “hosted” hypervisor _ install as an app on top of an existing OS _ performance penalty for the level of indirection _ e.g VMware Workstation, VMware Fusion, Microsoft Hyper-V, Parallels Desktop

Answer 51

_ on-demand _ self-service _ resource pooling _ elastic _ accessible _ measurable (charged for resources consumed) _ global reach (some sites are slow from other countries)

Answer 52

_ Amazon AWS _ Google GCP (Google Cloud Platform) _ Microsoft Azure

Answer 53

_ IaaS - infrastructure as a service _ PaaS - platform as a service _ SaaS - software as a service _ generically, XaaS, for anything shared as a service (directory service, backup, database -- DBaaS)

Answer 54

_ infrastructure as a service _ access to physical servers, storage, networking _ offers virtual machines, server virtualization, storage virtualization, network virtualization _ customer is responsible for the OS _ provider not responsible for consequences of hardware failure; e.g. customer is normally responsible for backups and redundancy

Answer 55

_ moving on-premise operations into the cloud | _ doesn’t take full advantage of the cloud, because it’s basically a transfer of local operations

Answer 56

_ virtual private cloud network | _ per customer, or to isolate testing from production

Answer 57

_ platform as a service _ the cloud provider provides virtual servers and services _ frees the developer from managing scaling and storage _ e.g. web hosting

Answer 58

_ software as a service | _ pretty much any network-based app

Answer 59

_ company owns and runs the cloud computing equipment, although it may be “co-located” (“co-lo”) in someone else’s data center _ sometimes requires for regulatory reasons, depending on what the company does

Answer 60

_ cloud provider company maintains the hardware for other companies _ the provider is sometimes able to provide better security than private customers can manage or afford _ multi-tenant

Answer 61

_ cloud jointly owned and operated by its tenants _ sometimes done by universities or governments _ cloud is “semi-private” for each tenant _ everyone shares maintenance and management

Answer 62

_ combination private/public cloud _ usually when a private company extends their in-house data centers via a public cloud _ useful for extra resources from public cloud when needed _ useful for isolating regulated functions in private cloud _ the private cloud connects to the public cloud via VPN (over the internet) or a telecom’s WAN

Answer 63

_ leveraging multiple cloud providers | _ done for different functions, for redundancy and flexibility, or to minimize expenses for certain functions

Networks Flashcards

(88 cards)