CS 372 Midterm
FTP is implemented over a pair of UDP connections. (T or F)
False
FTP is implemented over a single TCP connection. (T or F)
False
A server-side piece of data which is used to keep track of transactions between a client and server is called a cookie.
True
An FTP server maintains client states. (T or F)
True
HTTP implements a conditional GET using if-modified-since.
True
In store-and-forward transmission, a packet switch may begin transmitting the first bits of a packet before it has finished receiving that the last bits of the packet. ( True or False )
True
Viruses and spyware both infect a host computer by being (usually accidentally) downloaded. How is a computer virus different from spyware?
Viruses are designed to destroy files and cause a computer to malfunction in some way. Spyware is software that acts like a server. Once installed, a knowledgeable client can request and receive information from the infected host.
A network is:
a system for connecting computers using a single transmission technology
Consider the following: http://oregonstate.edu/admissions/sites/all/themes/doug_fir/images/osu-tag.png TLD:
edu
What are some application-layer protocols, and what are they used for? 3) SMTP/IMAP/POP3
As a packet is being constructed and passed "down" to the next layer of the internet protocol stack, a new "header" is added. This process is called:
encapsulation
Consider the following: http://oregonstate.edu/admissions/sites/all/themes/doug_fir/images/osu-tag.png URL name:
http://oregonstate.edu/admissions/sites/all/themes/doug_fir/images/osu-tag.png
Consider the following: http://www.google.com/images/srpr/logo11w.png The URL is:
http://www.google.com/images/srpr/logo11w.png
In a __________ architecture, one host is always on, and other hosts may connect and be handed off amongst themselves by this first host.
hybrid client-server/P2P
Protocol ________ allows multiple protocols to work together.
interoperability
How much longer does non-persistent HTTP take than persistent? (14 requests vs. 8 requests)
(14 req - 8 req) * 2 sec/req = 12 seconds. It takes 12 seconds longer
What are the major trade-offs implied by the use of cookies?
*more convenience for less privacy* Cookies make some web browsing more convenient by saving browsing history (e.g., shopping cart, location for weather report, etc.), but also allow servers to keep personal information about its clients, which may be seen as a violation of privacy.
Consider the following: http://oregonstate.edu/admissions/sites/all/themes/doug_fir/images/osu-tag.png Path name:
/admissions/sites/all/themes/doug_fir/images/osu-tag.png
Consider the following: http://www.google.com/images/srpr/logo11w.png The Path Name is:
/images/srpr/logo11w.png Answer 3:
The dotted-decimal form of 32-bit internet addresses is composed of 4 decimal numbers, separated by periods. What is range of possible values for each of the four decimal numbers?
0-255
Suppose that we send a DNS request with ID # 46921. What is the little-endian representation (hexadecimal)?
0x49B7
Suppose that we send a DNS request with ID # 46921. What is the big-endian representation (hexadecimal)?
0xB749
What are some tradeoffs for the implementation of RDT in TCP
1) Additional overhead 2) Possible delays due to waiting for acknowledgements 3) More packets on the network (because of ACKs) - leading to congestion issues 4) other answers possible...
How many unique network interface hardware addresses are possible?
12 hex digits means 16^12 = 281,474,976,710,656 unique numbers. Some of these are reserved.
A client's browser sends an HTTP request to a website. The website responds with a handshake and sets up a TCP connection. The connection setup takes 2 sec, including the RTT. The browser then sends the request for the website's index file. The index file references 6 additional image, which are to be requested/downloaded by the client's browser. How many requests (including the initial request) must be sent by the browser... a. With *non*-persistent HTTP? Show the Requests.
14 requests RQ #1: TCP Connection Request RQ #2: Website Index Request RQ #3: TCP Connection Request RQ #4: Image #1 Request ... RQ #13: TCP Connection Request RQ #14: Image #6 Request
The reserved port for FTP data transfer is port _____.
20
What is the total utilization of a circuit-switched network, accommodating 10 users with equal bandwidth, with the following users: • Four users are utilizing 100% their bandwidth • Two users are utilizing 60% of their bandwidth • Four users are inactive
4(1.00)(0.10) + 2(0.60)(0.10) + 4(0.00)(0.10) = 0.52 = 52%
The reserved port for FTP control is port _______.
21
A client in a network with a proxy server requests a 10 MiB file from an internet server, fakeservername.com. The network's proxy server has a 3.51 Mbps connection to fakeservername.com. The average response time between the network's proxy server and the internet origin server (including RTT) is 4.9 seconds for a small "header-only" HTTP request/response. The file requested by the client is currently in the proxy server cache, but the proxy server relays the client's request to the internet server with "if-modified since". Assume that transmissions between the proxy and the origin servers are stream (not packets) at full bandwidth, with negligible propagation delay. How much time is saved if the file has not been modified?
23.9 seconds
3 MiB to bits
25165824 bits
3 KiB to bytes
3072 bytes
A client's browser sends an HTTP request to a website. The website responds with a handshake and sets up a TCP connection. The connection setup takes 6.2 ms, including the RTT. The browser then sends the request for the website's index file. The index file references 5 additional images, which are to be requested/downloaded by the client's browser. Assuming all other conditions are equal, how much longer would non-persistent HTTP take than persistent HTTP? (Give answer in milliseconds, without units, rounded to one decimal place. For an answer of 0.01005 seconds, you would enter "10.1" without the quotes.)
31 milliseconds
How many unique 32-bit IP addresses are possible?
32 binary digits means 2^32 = 4,294,967,296 unique numbers. Many of these are reserved, so clearly there are not enough possible addresses for every individual host on the internet.
A client's browser sends an HTTP request to a website. The website responds with a handshake and sets up a TCP connection. The connection setup takes 2 sec, including the RTT. The browser then sends the request for the website's index file. The index file references 6 additional image, which are to be requested/downloaded by the client's browser. How many requests (including the initial request) must be sent by the browser... b. With persistent HTTP? Show the Requests.
8 requests RQ #1: TCP Connection Request RQ #2: Website Index Request RQ #3: Image #1 Request ... RQ #8: Image #6 Request
12 KiB to bits
98304 bits
What is the difference between the client/server model and the Peer-to-Peer (P2P) model
A client-server model has a dedicated service provider, and a dedicated service consumer. In a peer-to-peer model, there is no dedicated provider/consumer, and organization is handled amongst the hosts
What is an access network?
A network that allows a host to connect to the internet
What is RFC
A repository for comments, and the place to find (and post) the adopted standards for networking - in addition to a major resource for students and researchers in networking.
Place the steps in the correct order for a complete e-mail communication. Alice's user agent sends message to her mail server; message placed in message queue. Client side of SMTP opens TCP connection with Bob's mail server. Bob's mail server places the message in Bob's mailbox. Alice uses her user agent to compose message and sent to Bob's email address. SMTP client sends Alice's message over the TCP connection. Bob uses his user agent to read the message.
Alice uses her user agent to compose message and sent to Bob's email address. Alice's user agent sends message to her mail server; message placed in message queue. Client side of SMTP opens TCP connection with Bob's mail server. SMTP client sends Alice's message over the TCP connection. Bob's mail server places the message in Bob's mailbox. Bob uses his user agent to read the message.
What is "packet sniffing"?
An attacker gets access to communication hardware at some point in the route from a packet's source to its destination. By intercepting a packet and deconstructing it, the attacker can find out source/destination information, and can inspect the actual data message.
How does the receiver handle multiple incoming packets?
An input buffer is filled as data comes in, and the application layer drains the buffer as it needs the data contained therein.
What is the benefit of broadband cabling over baseband cabling?
Broadband cabling can utilize a greater range of frequencies in transferring information. This allows more data to be transferred over this type of cabling.
In the HTTP protocol, how is an object addressable?
By its Uniform Resource Locator (URL)
How does a client application communicate with a server application, after the connection has been established?
By sending data into a the socket involved in that connection
Is FTP a peer-to-peer or client/server protocol?
Client/Server
What are some human protocols?
Clinking a glass for a toast, tapping a glass with fork to announce a toast, raising your hand in a classroom, passing a "sharing stick" around a circle, Red/Yellow/Green traffic lights, etc.......
A client-side history of transactions between a client and server using HTTP is called a ___________.
Cookie
What internet service uses default ("well-known") port #606?
Cray Unified Resource Manager (URM)
What are some examples of guided media, and where are they used?
Cups with a string connecting them, twisted pair copper wires, coaxial cables, waveguide cabling (fiber-optic), etc... These are used in ADSL, cable internet
What states does the FTP protocol maintain?
Current Directory, Earlier Authentication, Limit on Concurrent Connections, etc...
An attack which attempts to make network resources unavailable to legitimate users by flooding a system with illegitimate traffic is called a ____ attack.
Denial of Service (DoS)
What are some responsibilities of the Application Layer?
Determine destination IP address Support network applications Decide which data which will transit the internet
If a botnet or another large group of hosts is used to execute this attack it is called a ________ attack
Distributed Denial of Service (DDoS)
What are some application-layer protocols, and what are they used for? 4) DNS
Domain name resolution (from www.google.com to 173.194.33.176, etc)
What are some applications which would have stringent security requirements?
Email, payment services, identity services
What devices reside on the network edge?
End Systems / Hosts (smartphones, computers, servers, peers, etc...)
What is the purpose of fast retransmit? How is it implemented?
Fast retransmit was originated to streamline the process of retransmissions, to increase utilization. If a TCP sender sees three duplicate ACKs (so four ACKs total) for the same segment, it assumes the segment was dropped, and retransmits, even if the countdown timer hasn't expired.
Application layer protocols must specify transport requirement by selecting a transport protocol from the transport layer. a) Give an example of an application that uses TCP, and explain why it uses TCP.
File downloading. TCP provides reliable transfer, so we can be confident that every bit of the file will be transferred and received in the correct order
How long does it take to send a 13 KiB file from Host A to Host B over a circuitswitched network, assuming: Total link transmission rate = 500Mbps The network is FDM 250 permitted users, each with an equal bandwidth share A link connection setup requires 100ms.
File size (L) = 13KiB = 13 × 1024 Bytes = 13 × 1024 × 8 bits = 106,496 bits One link's transmission rate (R) = 500Mbps / 250 users = 2Mbps = 2,000,000 bps Time to transmit file = L/R = 106496 bits / 2,000,000 bps = 0.053248 s = 53.25 ms Total sending time = setup time + transfer time = 100ms + 53.25ms = 153.25ms
What are some applications which would have stringent data integrity requirements?
File transfer, email, web services
What states does HTTP preserve?
HTTP is stateless. It does not preserve states.
What transport-layer protocol does HTTP use? What port number is reserved for this?
HTTP utilizes TCP, at port 80
What are some examples of unguided media, and where are they used?
Human speech, broadcast wireless, etc... Used in cellular networks, 802.11b/g/n
What is the minimum necessary information for identifying a process on a remote host? What do we call this identifier? What do we call a pair of these identifiers (one on each end host)
IP address & Port Number (process) Together, we call this a "socket" (or endpoint) We call a paired set of two sockets a "connection".
How is client-server architecture different from peer-to-peer (P2P) architecture?
In a client-server architecture, the server is always on and waiting for requests from clients, so when a request is received, the server can respond to the client. The client initiates communication with a server by sending a request.
What devices reside in the network core?
Interconnection devices (routers, switches, associated network cards, etc...)
Which are faster: Circuit-switched networks or Packet-switched networks?
It depends! A circuit-switched network may be faster due to the packet header overhead of packet-switched networks. However, a packet-switched network may be faster due to its more adaptive use of the transmission medium. It's even more complicated than this...
What are some applications which would have stringent timing requirements?
Video games, VoIP, streaming
If none of the links from the HTTP client to the HTTP server are congested, will a connection be faster with or without caching enabled.
It will still likely be faster with caching, because any time you request a file a second time, the proxy server will eliminate the necessity for the packets to traverse the internet, saving time.
What are the five layers in the Internet Protocol Stack?
Layer 1: Physical Layer Layer 2: Data-Link Layer Layer 3: Network Layer Layer 4: Transport Layer Layer 5: Application Layer
What are the seven layers in the ISO protocol stack? (Older, international version)
Layer 1: Physical Layer Layer 2: Link Layer Layer 3: Network Layer Layer 4: Transport Layer Layer 5: Session Layer Layer 6: Presentation Layer Layer 7: Application Layer
In terms of physical media, what is multiplexing
Merging multiple communication streams onto the same medium
How does SMTP send additional types of objects (other than raw text), since it isn't built to do so?
Multipurpose Internet Mail Extension (MIME) version inclusion in the header information allows the inclusion of additional data types.
A client's browser sends an HTTP request to a website. The website responds with a handshake and sets up a TCP connection. The connection setup takes 2 sec, including the RTT. The browser then sends the request for the website's index file. The index file references 9 additional images, which are to be requested/downloaded by the client's browser. How many requests (including the initial request) must be sent by the browser... Non-persistent HTTP requests: Persistent HTTP requests:
NP: 20 P: 11
Suppose that we send a DNS request with ID # 46921. Which representation is required for network communication?
Network order is big-endian, so 0xB749
For TCP, please select whether the service is implemented, and indicate how it is implemented. Bandwidth Guarantee?
No - no bandwidth guarantees.
For TCP, please select whether the service is implemented, and indicate how it is implemented. Jitter Threshold?
No - no jitter threshold.
Who invented the internet?
No one person invented the internet, but DARPA financed initial research, and a number of universities devoted significant research personnel to developing and establishing communication protocols for interconnectivity hardware and software.
What transport-layer services are provided by the base UDP protocol? (Choose all that apply) Minimum bandwidth Connection-oriented service Flow control Message security Congestion control Jitter minimization None of the above
None of the above
What are some applications which would have stringent bandwidth requirements?
Video games, large databases,
What's the difference between the POP3 and IMAP email protocols (from a user's point of view)?
POP3 has two possible modes: Mode (1) downloads emails to the client, and deletes them from the mail server Mode (2) downloads emails to the client, but keeps the originals on the mail server. IMAP displays emails on the client host, but maintains the client's entire email directory structure on the server. It also maintains the client's history (read/unread messages, etc.)
What are three important functions of a packet-switched network
Packet Construction, Transmission, and Interpretation
What happens if the sender sends information faster than the receiver can process?
Packets would be dropped. With a receive window implementation, the receive window size will shrink over time, eventually limiting the sender's output rate to that which can be processed by the receiver (flow control).
If an HTTP server can send 14 objects over a single TCP connection, this is an example of _____ HTTP.
Persistent
If a TCP sender sends packets with sequence numbers 100, 200, and 300, and has several more packets (all of 100 bytes) waiting in the sending buffer, and two ACKs are received back, with ACK numbers 200 and 300, what is the next step for the sender?
Re-transmit packet with sequence number 300.
What are the key points of a Selective Repeat retransmission scheme.
Selective repeat retransmission will keep a countdown timer for each un-ACK'd segment. When the protocol calls for a retransmission, it will only re-transmit un-ACK'd segments individually, rather than a bulk dump of all un-ACK'd segments.
If a TCP sender sends packets with sequence numbers 100, 200, and 300, and has several more packets (all of 100 bytes) waiting in the sending buffer, and only one ACK is received, with ACK number 400, what is the next step for the sender?
Send packet with sequence number 400
What transport-layer services are provided by the base TCP protocol? (Check all that apply) Reliable data transmission Jitter minimization Message security Congestion control Flow control Connection-oriented service Minimum bandwidth
Reliable data transmission Congestion control Flow control Connection-oriented service
Why are separate protocols required for sending email and receiving email?
Sending email requires a "push" protocol to send messages to the recipient's mail service provider. Receiving mail requires a "pull" protocol for the recipient to download messages from the mail service provider.
What is "IP spoofing"?
Sending information into the network with false "Sender IP" information.
How are these implemented in the HTTP handshaking procedure?
Server responds to client request, and includes in this response a request to use cookies. Client either sends next message with cookies enabled (to use cookies) or disabled. Sever generates cookie number and send this back to client.
Describe statistical multiplexing
Statistical multiplexing is the adaptive sharing of the transmission medium in packet-switched networks, based on whether or not hosts have data to transmit (on-demand), preventing the medium from being wasted by hosts without any data to transmit.
Application layer protocols must specify transport requirement by selecting a transport protocol from the transport layer. b) Give an example of an application that uses UDP, and explain why it uses UDP.
Streaming video. This requires fast data transfer, because video requires large amounts of data. UDP is much faster than TCP, because it does not perform all of the validation that TCP performs. While it is possible that some data may be lost or corrupted, it usually does not produce enough "jitter" to affect the performance.
In the internet, an application-level protocol implementing email service would most likely utilize _______ as its transport-layer protocol.
TCP
The FTP application-layer protocol utilizes the ___________ transport-layer protocol.
TCP
The HTTP application-layer protocol utilizes the ___________ transport-layer protocol.
TCP
What is the difference between TDM and FDM
TDM divides the available bandwidth into discrete blocks of time, during which only one of the users is permitted to transmit. FDM divides the available into bands of frequency (like radio stations), in which only one user is permitted to transmit.
What is the key point of a cumulative acknowledgement scheme?
The ACK you receive is for the next expected byte number. This means that an ACK for byte #N means that all bytes (N-1, N-2, ... 1) have already been received.
What is the Domain Name System (DNS) application-layer protocol used for? What transport-layer protocol does it make use of?
The DNS protocol uses UDP, and is used to resolve human-friendly word-based web addresses into Network-layer IP addresses and alias web and mail servers.
Why does the HTTP request message require an extra \r\n at the end of the header section?
The HTTP header must be separated from the entity body
ICANN (Internet Corporation for Assigned Names and Numbers)
The Internet Corporation for Assigned Names and Numbers is a multi-stakeholder group and nonprofit organization responsible for coordinating the maintenance and procedures of several databases related to the namespaces and numerical spaces of the Internet, ensuring the network's stable and secure operation.
What is the IETF
The Internet Engineering Task Force - who produces documentations which lay out standards for the Internet (without official decree).
InterNIC
The Network Information Center, also known as InterNIC from 1993 until 1998, was the organization primarily responsible for Domain Name System domain name allocations and X.500 directory services.
What organization manages the .org TLD?
The PIR, Public Internet Registry
A client in a network with a proxy server requests a 3MiB file from an internet server, x.y.z.com. The network's proxy server has a 1.54Mbps connection to x.y.z.com. The average response time between the network's proxy server and the internet origin server (including RTT) is 2 seconds for a small "header-only" HTTP request/response. The file requested by the client is currently in the proxy server cache, but the proxy server relays the client's request to the internet server with "if-modified since". Assume that transmissions between the proxy and the origin servers are stream (not packets) at full bandwidth, with negligible propagation delay. How much time is saved if the file has not been modified?
The connection and GET requests from the proxy server to the origin server, and the delivery from the proxy server to the client, are the same in either case. The only difference is in whether or not the file has been modified at the origin server, in which case the new file must be downloaded to the proxy server. As a stream download, this takes (3 x 1024 x 1024 x 8 bits) / (1,540,000 bits per second) ≈ 16.34 seconds.
What is the purpose of the TCP countdown timer?
The countdown timer is used in RDT to help determine if a packet has dropped. TCP will retransmit a packet, and restart the timer, upon a timer interrupt.
Why does FTP open two connections for a file transfer?
The first connection is for commands. This "command" connection allows the server to maintain the client state, so the client can traverse directories, etc., on the server. When a file download/upload is requested, the server opens a new connection that is exclusively for transferring the file. This "data" connection is closed when the file transfer is complete, but the "command" connection stays open until it is explicitly closed by either the client or the server.
What is done with network resources in a circuit switched network to allow multiple users to utilize the network?
The network resource is divided into chunks
What is the role of physical media in access networks.
The physical media is physical link in the chain between sender/receiver pairs. Analog signals relating to bits propagate over/through the physical media
In an internet name, what is the highest-priority component? The second-highest priority component? What are subsequently prioritized components used for?
The right-most (.com, .org) TLD is the highest priority. The next-leftmost ("Comcast" in www.comcast.net) is the domain name. fields to the left of this are used internally (such as "engr" in engr.oregonstate.edu) as host names.
HostA is sending fixed-size packets to remote HostB on a 100 Mbps link. Each packet's length is 1200 bytes. The propagation delay is 1 ms.; the processing/queuing delays are negligible. Given a sliding window of 12,000 bytes, and assuming no pipeline errors, what is HostA's utilization?
The time to transmit one packet is: L/R = (1200 x 8 bits) / 108 bps = 0.000096 sec = 0.096 ms. So it takes 0.96 ms to transmit 10 packets. The RTT is 2 ms. (i.e., propagation delay x 2) Total time before we can send packet # 11: 2.096 ms. Percentage of time actually transmitting: 0.96 / 2.096 = approximately 0.458 = 45.8%
___________ is the rate at which bits are actually transferred between sender/receiver.
Throughput
Transmisison delay
Time spent being placed on the transmission medium, limited by the transmission protocol. It is proportional to the rate at which the receiving node can interpret the electronic signal.
Processing delay
Time spent processing header information, checking bit errors, and deciding outport link
Propagation delay
Time spent traveling on the physical medium in between nodes. It is is proportional to the rate of travel of electrons through the transmission medium.
Queueing delay
Time waiting for the transmission medium
What are some services the TCP protocol does not provide
Timing, Bandwidth, (additional answers possible)
What is the goal of caching in HTTP?
To keep information requests from having to go to the goal server. This prevents the same information from being repeatedly downloaded, which would cause additional congestion in the internet (and slow down the rate at which the internet is perceived to load at the user end.
The _______ layer manages communications from process to process.
Transport
What must an application layer protocol specify?
Types of message exchanged Message Syntax (fields, structure) Message Semantics (meaning) Rules for message send & response (how, when)
UDP checksum
UDP checksum is a the complement of a 16-bit one's complement sum calculated over an IP "pseudo-header" and the actual UDP data. The IP pseudo-header is the source address, destination address, protocol (padded with a zero byte) and UDP length.
What are some application-layer protocols, and what are they used for? 1) HTTP
Used often for communication of hosted information on the internet (web pages)
What are some application-layer protocols, and what are they used for? 2) FTP
Used often for transfer of files without a web interface (uploading, etc)
What are the three major components of e-mail?
User agent (such as an email client, or webmail); a mail server; an email protocol.
What are some application-layer protocols, and what are they used for? 5) SCCP
VOIP
If a TCP sender sends packets with sequence numbers 100, 200, and 300, and has several more packets (all of 100 bytes) waiting in the sending buffer, and no ACKs are received back, what is the next step for a TCP sender?
Wait for a timer interrupt. If countdown timer expires, re-send packet with sequence number 100.
Is 130.154.1.25 a valid IP address?
Yes
For TCP, please select whether the service is implemented, and indicate how it is implemented. Reliable Delivery(mostly)?
Yes - Acknowledgement for delivery of packets
For TCP, please select whether the service is implemented, and indicate how it is implemented. Full Duplex Service?
Yes, for example Telnet, which echoes back every character.
For TCP, please select whether the service is implemented, and indicate how it is implemented. Connection-Oriented?
Yes, implemented by a 3-way handshake at start of connection.
For TCP, please select whether the service is implemented, and indicate how it is implemented. Congestion Control?
Yes, implemented by restricting pipeline size based on receipt of acknowledgements (or lack thereof)
For TCP, please select whether the service is implemented, and indicate how it is implemented. Flow Control?
Yes, receiver advertises a window size, and sender limits output accordingly
For TCP, please select whether the service is implemented, and indicate how it is implemented. Pipelining?
Yes, though pipeline is limited by congestion and flow control
For TCP, please select whether the service is implemented, and indicate how it is implemented. In-order delivery?
Yes, via sequence numbers in TCP header information
The internet is:
a network of networks
With a stop-and-wait implementation a. How long would it take to send five 1500-byte packets, subject to the following conditions: • R = 20Mbps • Network end-to-end time (after initial transmission) = 20 ms • ACK packet L = 20 bytes b. What is the utilization?
a) • Initial Transmit = 1500 * 8 / 20,000,000 = 0.6 ms. • Propagation = 20 ms • ACK transmit = 20 * 8 / 20,000,000 = 0.008 ms • Propagation = 20ms • First packet total = 0.6 + 20 + 0.008 + 20 = 40.608ms (Just the first packet) • Repeat 5x = 203.04ms b) Time spent transmitting = 5*(0.6 + 0.008) = 3.04ms Total time until finished = 203.04ms (from above) 3.04ms / 203.04ms = approx. 1.5 %
Consider two hosts (A and B) connected by a single link with transmission rate R bps. The hosts are separated physically by d meters along a cable for which the speed of propagation is s meters per second. Host A sends a packet of length L bits to host B a. Show the propagation delay d.prop in terms of d and s: b. Show the transmission delay d.trans in terms of L and R: c. Suppose Host A begins to transmit at time t = 0. At time t = d.trans where is the last bit of the packet? d. Suppose that dprop > dtrans . At time t = dtrans , where is the first bit of the packet? e. Suppose that dprop < dtrans . At time t = dtrans , where is the first bit of the packet? f. Let s = 2.5x108 m/s, L = 120 bits, R = 56 Kbps. Find d such that d.prop = d.trans.
a. d.prop = d/s b. d.trans = L/R c. Just leaving the sending host (Host A) d. Somewhere in the link between Host A and Host B e. At the receiving host (Host B) f. Set d/s = L/R 120 / 56,000 = d / 250,000,000 d = 250,000,000 * 120 / 56,000 = 535,714.28 m ≈ 535.7 km
The payload (non-header portion) of a transport-layer segment is
application data
A _______ is a group of computers all controlled for a common objective, possibly by an infection in the system, used cooperatively to process and/or send information in mass amounts (DDoS, distributed computing)
botnet
What units are used for transmission rate?
bps
What units are we using for network bandwidth?
bps
A ________-switched network has a dedicated circuit through which information is sent, whereas a ________-switched network sends information in "chunks" through a network on shared media
circuit packet
Consider the following: http://www.google.com/images/srpr/logo11w.png The TLD is:
com
What is the equation for end-to-end delay?
d.nodal = d.proc + d.queue + d.trans+ d.prop
Assuming that all other conditions are equal, which type of HTTP (persistent, non-persistent) takes longer to complete the entire transfer?
non-persistent
Consider the following: http://oregonstate.edu/admissions/sites/all/themes/doug_fir/images/osu-tag.png Host name:
oregonstate.edu
The internet core is a ____________ switched network.
packet
If an HTTP server can send 2 objects over a single TCP connection, this is an example of ____________ HTTP.
persistent
We call a program that is running on a host a _______.
process
IMAP is a ________ protocol
pull
POP3 is a ________ protocol.
pull
SMTP is a ________ protocol
push
A paired IP address and port number is called a:
socket
Two Generals' Problem
the Two Generals' Problem is a thought experiment meant to illustrate the pitfalls and design challenges of attempting to coordinate an action by communicating over an unreliable link. In the experiment, two generals are only able to communicate with one another by sending a messenger through enemy territory. The experiment asks how they might reach an agreement on the time to launch an attack, while knowing that any messenger they send could be captured.
Consider the following: http://www.google.com/images/srpr/logo11w.png The Host Name is:
www.google.com
What are some components of the TCP protocol?
• Connection-Oriented with 3-way connection-setup handshake • Reliable, in-order data transfer • Flow control • Congestion Control
What are some components of the UDP protocol?
• Connection-less • Unreliable data transfer (Fire & Forget) • No flow or congestion control
Applications that use TCP:
• HTTP (Web) • FTP (file transfer) • Telnet (remote login) • SMTP (email)
What are some of the effects of congestion in a packet-switched network?
• Packet Delay • Packet Loss • Jitter • etc..
What are some reasons for the layering of network protocols?
• Protocols can be tested independently of one another • When maintenance is required at one level, changes do not affect other layers • Can update the inner workings of a protocol, as long as input/output remain the same • The complication of dealing with the intermeshed types of hosts and data is lessened
What does a protocol do?
• Specifies the format and order of messages sent and received among network entities. • Specifies actions taken upon the transmission and receipt of messages
What are some factors in why computer networking is so complicated?
• Variety of hardware devices which must communicate with each other (legacy, modern) • Software being developed which must be interoperable • Numerous protocols and communication standards • Complicated terminology and acronyms • Wireless and Mobility challenges • etc...
Communication via guided media is complex enough by itself, but communication via unguided (wireless) media is even more complicated. What are some additional difficulties of wireless networking?
• Wireless signals are more vulnerable to interference and noise. • Wireless signals may be blocked, especially by metal. • Wireless signals may reflect and interfere with other wireless signals.
Applications that use UDP
• streaming media • Teleconferencing • DNS (Domain Name Service) • Internet telephony