Computer Science 360 Quiz 1 Name___________________ Rules: 1 - Closed book, notes, neighbor's paper 2 - Please write your name on front AND back of paper 3 - DON"T fold your paper 4 - You may leave when you finish 1. Rank the following transmission media from faster (1) to slower (2). ____ a. Wireless radio ____ b. Wired ethernet 2. Rank the following transmission media from highest bit error rate (1) to lower bit error rate (2). ____ a. Wireless radio ____ b. Wired ethernet 3. Which of the following layers in the ISO/OSI model is present in hosts but not in routers and switches a. physical layer b. transport c. network d. data link 4. A packet being transmitted is said to travel down the protocol stack toward the physical layer. As this occurs: a. headers are removed from the b. existing headers are overwritten packet c. headers are appended to the d. headers are prepended to the end of the packet front of the packet 5. Which layer is responsible for routing a packet toward its destination a. transport b. data link c. network d. session Computer Science 360 Quiz 2 Name___________________ 1. Which of the following is NOT a layer in the TCP/IP layered architecture a. transport b. link c. network d. session 2. The UDP protocol resides at what layer a. transport b. link c. network d. session 3. Multibyte addresses must be specified in a. big endian format b. little endian format (most significant byte first) (LSB first) c. the format used by the source d. the format used by the computer destination computer 4. Which of the following addresses must be specified by the APPLICATION that wishes to send a packet over an IP network (circle all that apply) a. destination MAC address b. destination IP address c. destination transport protocol ID d. destination Port address 5. Which of the following addresses is used in routing an IP packet sent from the US to India as it passes though intermediate routers a. destination MAC address b. destination IP address c. destination transport protocol ID d. destination Port address Computer Science 360 Quiz 3 Name___________________ 1. The socket API used in network programming employs functions from a. the low level I/O library b. the standard I/O library c. functions from either library d. the socket API has no relation may be used interchangably to either Input/Output library 2. In the socket API the function that can be used to "automagically" print a descriptive message when a function fails is: a. fprintf(stderr, "%d", errno); b. werror(); c. perror(); d. errno_print(); 3. On success the socket() function returns a. a FILE * pointer b. a small integer c. a SOCK * pointer d. -1 4. To create a UDP socket the SECOND of the three parameters passed to socket() should be: a. SOCK_DGRAM b. AF_INET or PF_INET c. IPPROTO_IP d. SOCK_UDP 5. The bind function is used to a. bind a socket to a remote b. bind a socket to a local port and IP address port and IP address c. bind is used to perform both functions Computer Science 360 Quiz 4 Name___________________ 1. Which of the following best characterizes the way the DNS presently works a. every week or so a central server b. every week or so sysadmins downloads a database that maps send an email containing all all names on the Internet to their of their host names and IP IP address to each host on the addresses to a central DNS Internet administrator who updates a global data base referenced by all "root name servers" c. every week or so hosts on the d. The DNS system is a heirarcical Internet upload their own names and distributed database and there IP addresses to a central server does not exist ANY central that provides name service to the repository mapping all names entire internet. to IP address. 2. When a UDP (SOCK_DGRAM, IPPROTO_IP) socket is in use, an attempt to connect() the socket a. will fail unless an application b. will fail even if an application on the remote host has bound to on the remote host is bound to the remote port the remote port UNLESS the process bound to the socket is trying to read from it. c. will succeed ONLY if the remote d. can return success even if the host address in the struct remote IP address is not sockaddr_in specifies the IP address presently assigned to ANY host. of a host on the Internet that is presently powered on. 3. Root nameservers will communicate a. only with authorized nameservers b. only with applications running with root (superuser) privileges c. with any program regardless of whether it is an "official" nameserver or its privilege state 4. Before calling gethostbyname() a. a socket MUST be created b. a socket must NOT be created c. it doesn't matter whether or not a socket has been created. 5. Suppose a makefile contains the following lines: connect: connect.c ${DEPS} gcc -o $@ $@.c ${DEBUG} ${OBJS} 2> $@.err when the user enters "make connect" the value of $@ becomes: a. a.out b. connect.o c. connect: d. connect Computer Science 360 Quiz 5 Name___________________ 1. An application protocol that uses UDP a. must supply a full and correct b. must supply a UDP header UDP header in the data passed but the UDP layer will by write() or sendto() compute the checksum c. Cannot supply the UDP header as it is fully built by the UDP layer. 2. If a UDP application sends a 40 byte datagram to a port on a remote host and no application is bound to the remote port. a. sendto() will return -1 b. sendto() will still return instead of 100 100 and the application will receive no indication the datagram was discarded. c. sendto will return 100 and, d. the application will receive on the next read from the a segfault signal and will socket the application will terminate inside the call to receive an icmp error message sendto() 3. Which best characterizes the use of write() and sendto() in sending UDP data. a. The socket must be connected b. The socket must be connected before write() can be used. before sendto() can be use c. The socket must be connected d. The socket need not be connected before BOTH can be used. before either is used. 4. On a Linux system, if a UDP packet contains 30 bytes of application data and the receiver calls: amt = read(sock, buf, 10); a. The first 10 bytes of the packet b. All 30 bytes will be copied will be put in the buffer and to location buffer potentially the remainder discarded. overwriting other data in the program c. The first 10 bytes of the packet d. read() will return -1 will be put in the buffer and the remainder saved for subsequent reads() 5. Suppose a UDP receiver wishes to know the IP address and port of the sender. a. It must use read() b. It must use recvfrom() c. It may use either one. d. There is no way to find out this information! Computer Science 360 Quiz 6 Name___________________ 1. When a program calls pthread_create() to create a new thread it must provide: a. the name of an executable b. the address of a function program such as "vi", "hostinfo", contained in the program or "a.out" c. the address of any line of d. all of the above must be code in the program. provided 2. In the design of the udping program which of the following involved the use of two threads (circle all that apply a. the udping server code b. the client sender c. the client receiver d. all three needed two threads 3. The function that should be used to retrieve the current time of day at near microsecond resolution is: a. time() b. pthread_cond_timedwait() c. gettimeval() d. gettimeofday() 4. Suppose the following option string is passed to getopt() char *optargs="icp:s:Sn"; For which set of options are additional arguments expected to be present: a. -i -c -p b. -s -S c. -p -s d. -i -c -n 5. Which best distinguishes the struct timeval and the struct timespec a. timeval is double precision but b. timeval is in seconds and timespec is in seconds and micro-seconds but timespec nano-seconds is double precision c. timeval is in seconds and nano- c. timespec is in seconds and nano- seconds but timespec is in seconds but timeval is in seconds and micro-seconds seconds and micro-seconds Computer Science 360 Quiz 7 Name___________________ 1. When a program calls signal() to set up a signal handler it must provide: a. the name of an executable b. the address of a function program such as "vi", "hostinfo", contained in the program or "a.out" that will handle the signal. c. the address of any line of d. all of the above must be code in the program. provided 2. The process of preventing a sender from generating packets faster than the receiver is consuming them is known as a. congestion control b. flow control c. connection management d. traffic squelching 3. The process of preventing a sender from generating packets faster than the network can deliver them is known as a. congestion control b. flow control c. connection management d. traffic squelching 4. Of the two processes described in questions (2) and (3) which best describes the difficulty of implementing them. a. both are pretty much b. (2) is straightforward impossible but (3) is more difficult c. both are trivial d. (3) is straightforward but (2) is more difficult 5. Consider a protocol in which a special ACK packet exists but no sequence numbers are actually transmitted. The Sender sends a packet and then waits for an ACK packet or a timeout. If a timeout occurs the sender retransmits the last packet sent. Which best describes this protocol's operation. a. if an ACK is dropped by the b. if a packet is dropped network the previous packet by the network the may be lost forever. receiver will end up with a duplicate copy of the packet c. if an ACK is dropped d. if a packet is dropped by the network the by the network the receiver may end up with packet will not be resent a duplicate copy of a packet and thus lost forever Computer Science 360 Quiz 8 Name___________________ Suppose a sender sends a burst of 5 packets having sequence numbers {0, 1, 2, 3, 4} using a go-back-n sliding window protocol and the packet with sequence number 2 is lost by the network. Assume NO timeout and re-transmissions have occurred YET. 1. After the receiver receives the packet with sequence number 4, its value of nxtrcv will be: a. 4 b. 5 c. 3 d. 2 2. After the receiver receives the packet with sequence number 4, it will send back an ACK packet having ACK number: a. 4 b. 5 c. 3 d. 2 3. After the sender receives the ACK mentioned in question 2 its value of lstack will be: a. 4 b. 5 c. 3 d. 2 Now assume that the sender timeout routine recognizes that an ACK is way overdue. 4. The sender will retransmit packets having sequence numbers a. 0,1,2,3,4 b. 0 only c. 2, 3, 4 d. 2 only 5. Suppose a Stop and Wait protocol (in which the sender can send only ONE packet and then must wait for an acknowledgement) is in use. Suppose the round trip delay from the time the packet is sent until the ACK is received is 50msec. Suppose each packet carries 10,000 bits. Suppose the bit rate of each link on the path is 1,000,000,000 bits per second. The maximum data rate this protocol can achieve is: a. 10,000 bits/sec b. 100,000 bits/sec c. 200,000 bits/sec d. 1,000,000,000 bits/sec 6. In my previous CPSC courses I had: a. java in 101, 102, 212 b. C in 101, 102, 212 C in 215 Java in 215 c. Other (identify) Computer Science 360 Quiz 9 Name___________________ 1. The recommended maximum on the number of lines of code that should be written before a compile is: a. 1 b. 10 c. 100 d. 1000 2. Writing code that looks like: count = (a->b[myfun.sub]->start++) + --j*bigval(testit(x % 15)/ 3 * y); a. Should be done whenever possible b. Is a GREAT way to convince your because it greatly speeds execution coworkers how brilliant you are. times c. Is a good way to convince knowledgable coworkers what an incompetent wannabe you are. 3. The consequence of using the traits of an ineffective programmer is: a. it takes you longer to get the b. it takes less time to get the program entered into the computer program entered but debugging but debugging is way faster. is way slower. c. both program entry and debugging are d. both program entry and debugging way faster. are way slower. 4. The number of lines of debugging diagnostics per line of real code used by an inexperienced programmer in developing his or her first multithreaded program is closest to: a. one line per function b. one line per every 10 lines of code. c. one line per line of code 5. The effect of moving the comment // from the second line to the first line is: #define dprintf(format, args...) fprintf(stderr, format, ##args) // #define dprintf(format, args...) a. disable ALL dprint() diagnostics b. disable ONE dprintf() a. enable ALL dprint() diagnostics b. enable ONE dprintf() Computer Science 360 Quiz A Name___________________ NOTE: p_m_l() = pthread_mutex_lock() p_m_u() = pthread_mutex_unlock() p_c_w() = pthread_cond_wait() p_c_s() = pthread_cond_signal() 1. Suppose a function needs to wait until the value of "items" is non-zero. For correct operation the function must lock the mutex associated with "items" a. before it tests "items" or b. only before it calls p_c_w() calls p_c_w() c. Its not necessary to lock the mutex at all. 2. Suppose another function is going to increment items and signal the cv. For correct operation the function must lock the mutex associated with "items" a. before it increments "items" or b. only before it calls p_c_s() calls p_c_s() c. Because signaling never causes a thread to block its never necessary to lock mutex at all. 3. Which of the following strategies are correct for the signaller (circle all that are). a. items++; b. p_m_l(&item_mtx); p_m_l(&item_mtx); p_c_s(&item_cv); p_c_s(&item_cv); p_m_u(&item_mtx); p_m_u(&item_mtx); items++; c. p_m_l(&item_mtx); d. p_m_l(&item_mtx); p_c_s(&item_cv); items++; items++; p_c_s(&item_cv); p_m_u(&item_mtx); p_m_u(&item_mtx); 4. The RPS CONNECT packet type is legal only in which of the following states: a. RPS_CLOSE b. RPS_LISTEN c. RPS_CONN_PEND d. RPS_CONN_ACK 5. The RPS CONNECT ACK packet type is legal only in which of the following states: a. RPS_CLOSE b. RPS_LISTEN c. RPS_CONN_PEND d. RPS_CONN_ACK Computer Science 360 Quiz B Name___________________ NOTE: p_m_l() = pthread_mutex_lock() p_m_u() = pthread_mutex_unlock() p_c_w() = pthread_cond_wait() p_c_s() = pthread_cond_signal() 1. In an "infinite" ring buffer which of the following is true: a. nextin >= nextout b. nextin <= nextout c. depending on circumstances d. they are always equal. sometimes nextin > nextout and sometimes nextout > nextin 2. In a producer/consumer problem the producer calls a. ring_get_item() and b. ring_get_slot() and ring_add_item() ring_add_slot() a. ring_get_item() and b. ring_get_slot() and ring_add_slot() ring_add_item() 3. In a producer/consumer problem the consumer calls a. ring_get_item() and b. ring_get_slot() and ring_add_item() ring_add_slot() a. ring_get_item() and b. ring_get_slot() and ring_add_slot() ring_add_item() 4. The ring_add_slot() function must invoke the following pthread calls: (circle all that are true) a. p_m_l() b. p_m_u() c. p_c_w() d. p_c_s() 5. The ring_get_item() function must invoke the following pthread calls: (circle all that are true) a. p_m_l() b. p_m_u() c. p_c_w() d. p_c_s() Computer Science 360 Quiz C Name___________________ The following questions assume a simplex connection in which one endpoint acts as a sender of data and the other acts as receiver. 1. The seq number carried in a packet will a. increase on packets going from b. increase on packets going from sender to receiver but stay the receiver to sender but stay the same on packets going from same on packets going from receiver to sender sender to receiver c. increase on packets flowing in d. Not change on packets flowing both directions in both directions. 2. The ack number carried in a packet will a. increase on packets going from b. increase on packets going from sender to receiver but stay the receiver to sender but stay the same on packets going from same on packets going from receiver to sender sender to receiver c. increase on packets flowing in d. Not change on packets flowing both directions in both directions. 3. Slots in the sender send ring will be returned to the free slot set a. as soon as the packet has been b. during ack processing by the transmitted by the sender thread sender thread. c. during ack processing by the d. during rps_write() calls in the receiver thread main thread. 4. Suppose an ACK packet arrives at the sender with ACK=7,WIN=5 a. How many free slots were there in the receiver's receive ring at the time the packet was sent? b. When the sender receives the ACK suppose nxtsnd=9. How many MORE packets can't be sent before the sender must stop and wait for a new ack. Computer Science 360 Quiz D Name___________________ For each of the operations described below identify the component of the rps system that carries out the operation. ____ 1. Invoke ring_get_slot() on the a. The rps_write() API function sender ring b. The rps_read() API function ____ 2. Invoke ring_add_slot() on the sender ring c. The rps_sender() thread ____ 3. Invoke ring_get_item() on the d. The rps_receiver() thread sender ring ____ 4. Invoke ring_add_item() on the sender ring. ____ 5. Invoke ring_get_slot() on the receiver ring ____ 6. Invoke ring_add_slot() on the receiver ring ____ 7. Invoke ring_get_item() on the receiver ring ____ 8. Invoke ring_add_item() on the receiver ring. 9. Which of the following should NEVER cause the caller to have to actually wait on a condition variable. a. Invoke ring_get_slot() on the b. Invoke ring_get_slot() on the sender ring receiver ring c. Invoke ring_get_item() on the b. Invoke ring_get_item() on the sender ring receiver ring Computer Science 360 Quiz E Name___________________ 1. The function call: write(sock, buffer, 1000000) a. will not work with both TCP and UDP b. works for TCP but not UDP sockets c. works for UDP but not TCP d. works for both TCP and UDP 2. When an out of order packet is received, it is a. buffered by both TCP and RPS b. buffered by RPS but discarded by RPS c. buffered by TCP but discarded d. discarded by both RPS and TCP by RPS 3. For a starting sequence number of a connection a. both RPS and TCP use 0 b. RPS uses 0 but TCP does not c. TCP uses 0 but RPS does not. d. Both RPS and TCP derive the starting sequence number from a clock 4. Each time a data packet is sent a. the seq number of the next packet b. the seq number of the next packet increases by 1 in both RPS and TCP increases by 1 in TCP but by the amount of data in the previous packet in RPS c. the seq number of the next packet d. the seq number of the next packet increases by the amount of data increases by 1 in RPS but by the in the previous packet in both amount of data in the previous TCP and RPS packet in TCP 5. RPS has greater robustness in the face of delayed packets than does TCP because of a. the sig field in the header b. the magic number field c. both a and b are equally responsible. Computer Science 360 Quiz F Name___________________ Suppose a TCP 3 way handshake is being used to establish a connection. Let SYN1 represent a packet that carries SYN and no ack. Let SYN2 represent a packet that carries both SYN and ACK that is sent in response to a SYN1 1. A old delayed SYN2 received in the SYN sent state is: a. Detected because the b. Detected because SYN2 is SEQ number is wrong. unexpected in the SYN sent state c. Detected because the d. Undetected and responded to ACK number is wrong. with an SYN2 2. A old delayed SYN1 received in the LISTEN state is: a. Detected because the b. Detected because SYN1 is SEQ number is wrong. unexpected in the LISTEN state c. Detected because the d. Undetected and responded to ACK number is wrong. with an SYN2 3. A old delayed SYN1 received in the ESTABLISHED state is: a. Responded to with RESET. b. Responded to with an ACK packet specifying current ACK and SEQ. c. Responded to with SYN2 d. Responded to with SYN1. 4. Which of the zones in in the diagram represents data that CAN be copied to TCP send buffers but CAN"T actually be sent yet: 1 2 3 4 5 ----------|----------|----------|--------|------- SND.UNA SND.NXT SND.UNA SND.UNA +SND.WND +SND.BUF a. 2 b. 3 c. 4 d. 5 5. Which of the zones in the diagram represents that data that has been ACKED but not yet consumed by the application. 1 2 3 4 ----------|----------|--------|--------- RCV.APP RCV.NXT RCV.APP +RCV.BUF a. 1 b. 2 c. 3 d. 4 Computer Science 360 Quiz G Name___________________ 1. Suppose receive buffer space is 10,000 bytes, rcv.app = 4000 and rcv.nxt is 7000. a. What values will be sent in outgoing packets for WINDOW = ACK = b. Supose a packet with SEQ = 9000 and length = 1000 is now received. What will outgoing window and ack be now? WINDOW = ACK = c. Suppose a packet with SEQ = 7000 and length = 2000 is now received What will outgoing window and ack be now? WINDOW = ACK = d. Finally assume that the application reads 5000 bytes. What will outgoing window and ack be now? WINDOW = ACK = 2. Suppose transmit buffer space is 20,000, snd.ntx = 24,000 snd.una = 22,000 and a packet is received with ACK = 26000 and WINDOW = 12000. After the incoming packet is processed, a. Suppose that the full complement of bytes are presently buffered, how many bytes may the sending TCP send before it must stop sending. Computer Science 360 Quiz H Name___________________ 1. Suppose a link can carry 20,000,000 BYTES per second and that the round trip time is 0.01 seconds. Suppose packet size is 2000 bytes. What is the smallest sender window size in PACKETS that will permit sending without stopping in the absence of errors. a. 20,000,000 * 2000 * 0.01 b. 20,000,000 / (0.01 * 2000) c. 20,000,000 * 0.01 / 2000 d. 20,000,000 * 2000 / 0.01 2. Using the bandwidth delay product to determine sender window size a. is NEVER appropriate because b. is appropriate for link layer it causes congestion protocols on point to point links but not for TCP. c. is not appropriate for link d. Works well for both link layer layer protocols but is OK for protocols and for TCP TCP 3. Suppose packets are 10,000 bits in length and the bit rate on the outgoing link is 1,000,000 bits / second. What will be the queuing delay IN SECONDS of a packet that enters output queue with 200 packets in line ahead of it? a. 200 * 10000 * 1,000,000 b. 10000 * 200 / 1,000,000 c. 1,000,000 / (200 * 10000) d. 1,000,000 * 200 / 10000 4. During the slow-start procedure the size of the cwnd a. increases by one each round b. doubles each round trip time trip time. c. doubles each time an ACK is received 5. Before a TCP sender can send a packet it must be in: a. JUST the congestion window b. BOTH the congestion window and or the usable window the usable window. c. JUST the congestion window d. JUST the usable window. Computer Science 360 Quiz I Name___________________ Suppose MSS = 1000 bytes, cwnd = 10000, snd.seq = 18000, snd.una = 12000 and the offered window is 32000. 1. If the segment that has sequence 12000 was actually lost, give the starting sequence numbers of the remaining 1000 byte segments that the sender can send before having to stop. (Hint: all sequence numbers end in 000) 2. Suppose that the sender has stopped when the 3rd duplicate ack arrives. What does the receipt of the third duplicate ack cause ssthresh and cwnd be set to (in BYTES) ssthresh = cwnd = 3. How many additional duplicate acks will have to be received before a NEW segment can be sent. 4. Assuming that ONLY the 12000 segment was lost what will be the first ACK for new data that is received) (Hint: the answer ends in 000) 5. What will be the leading edge of the congestion window when the first ack for new data is received (usual hint). Computer Science 360 Quiz J Name___________________ 1. The rps_drop_pkt() routine decides whether or not a packet should be dropped based upon. a. a call to the random() b. The low order byte of the seconds random number generator component of the time of day. c. The high order byte of the micro d. The low order byte of the micro seconds component of the time of second component of the time of day. day 2. Which of the following is the timestamp that should be used in checking to see if a packet timeout has occured. a. rpsock->sndring.bufs[ rpsock->lstack].ts; b. rpsock->sndring.bufs[ rpsock->sndring.nextout].ts; c. rpsock->sndring.bufs[ rpsock->nxtrcv].ts; d. rpsock->sndring.bufs[ rpsock->sndring.nextin].ts; 3. Which of the following conditions MUST hold before a RPH_WUP control packet is generated. a. the usable window is 0 b. the offered window is 0. c. the number of free slots in d. there are no unacked packets. receive ring is 0 4. The rps_go_backn() function retransmits all packets whose sequence numbers are: a. > lstack and < nxtsnd b. >= lstack and < nxtsnd c. > lstack and <= nxtsnd d. >= lstack and >= nxtsnd 5. Which of the following causes the rps_close() procedure to be UNable to complete normally: a. The loss of the RPH_CLOSE b. The loss of the RPH_CLOSE | control packet RPH_ACK. c. The loss of either one Computer Science 360 Quiz K Name___________________ +---------+ snd FIN / \ snd ACK +---------+ | FIN |<----------------- ------------------>| CLOSE | | WAIT-1 |------------------ | WAIT | +---------+ rcv FIN \ +---------+ | rcv ACK of FIN ------- | CLOSE | | -------------- snd ACK | ------- | V x V snd FIN V +---------+ +---------+ +---------+ |FINWAIT-2| | CLOSING | | LAST-ACK| +---------+ +---------+ +---------+ | rcv ACK of FIN | rcv ACK of FIN | | rcv FIN -------------- | Timeout=2MSL -------------- | | ------- x V ------------ x V \ snd ACK +---------+delete TCB +---------+ ------------------------>|TIME WAIT|------------------>| CLOSED | +---------+ +---------+ Label each of the following states T (transient) or P(persistent) depending upon whether or not the state lasts only the time it takes to send a packet and receive an ACK (T) or whether the state can persist for an arbitrarily long time ____ 1. Fin Wait-1 ____ 2. Fin Wait-2 ____ 3. Close Wait ____ 4. Last ACK Label each of the following socket calls B (blocking) or N (non-blocking) depending upon whether return to the caller occurs within milliseconds (N) or may take an arbitrarily long amount of time: ____ 5. listen() ___ 6. accept() 7. It is common for multiple threads of a single server to be bound to a. the same socket and port b. the same socket but different ports c. different sockets but the same port. 8. For delivery of incoming TCP packets to the proper socket is based upon it suffices to know a. the destination port b. the destination port and IP address c. source and destination ports d. source and destination port. and IP addresses Computer Science 360 Quiz L Name___________________ 1. Which best describes how fragmentation in IP works a. packets are always fragmented by b. packets are fragmented by the the sender and reassembled by the "bottleneck" router and receiver reassembled by the receiver c. packets are fragmented by the "bottleneck" router and reassembled as soon as they reach a link that can hold them. 2. Fragments can be correctly mapped to the proper reassembly queue using a. only the ID number in the b. using only the source/dest IP IP header addresses c. both (a) and (b) are required 3. Suppose the Charter cable company is assigned a 20 bit prefix for one of its local operations. The number of possible host addresses that can be associated with the 20 bit IP prefix is closest to: a. 12 b. 2^20 c. 2^12 d. 2^32 4. To which of the old classful addresses (A, B, or C) does each of the following addresses belong. ____ a. 188.30.254.1 ____ b. 124.1.15.99 5. The number of hosts that can be assigned to a subnet with mask 255.255.255.192 is closest to: a. 192 b. 256 c. 16 d. 64 Computer Science 360 Quiz M Name___________________ 1. Suppose a prefix is 5 bits long. The "prefix mask" used in prefix based routing is: a. 00000111111111111111111111111111 b. 11111000000000000000000000000000 c. 11111111111111111111111111100000 d. 11111111111111111111111111100000 2. Before comparing a destination address with the prefix it is necessary to a. OR it with the prefix mask b. XOR it with the prefix mask c. AND it with the prefix mask d. Any of the a, b, or c will work. 3. Suppose a prefix is 130.126/15 which of the following addresses match the prefix (Circle all that do) a. 130.125.0.1 b. 130.126.0.1 c. 130.127.0.1 d. 130.128.0.1 4. Which of the following MUST be transmitted via link level broadcast? a. IP packet b. Arp request c. Arp reply d. both b. and c. 5. For efficiency in routing table look ups its a good idea to order the routing table from: a. longest prefix to shortest b. shortest prefix to longest c. in increasing prefix value d. randomly Computer Science 360 Quiz N Name___________________ 1. Let N = number of nodes in a network and M = the average number of links / node. a. In terms of computational complexity the OSPF algorithm as performed by a single node is of order: 1. M 2. N 3. N ^ 2 4. M ^ 2 5. N ^ M b. The minimum number of routing packets transmissions at each complete exchange in an OSPF network within the entire network is of order: 1. M 2. N 3. N ^ 2 4. M ^ 2 5. N * M 2. Each individual routing update packet propagates a. Across the entire net in OSPF b. Only to adjacent neighbors in OSPF and in distance vector and distance vector c. Across entire net in OSPF but d. Across entire net in distance vector only to adjacent neighbors inF but only to adjacent neighbors distance vector in OSPF. 3. The routing algorithm used to transmit routing updates in an OSPF network is typically: a. Static tables built by b. random routing sysadmin c. distance vector d. Flooding 4. Suppose the distance vector routing algorithm is being used and station B receives the following routing updates from neighbors A, C and E Suppose BA delay = 2, BC delay = 4 and BE delay = 3. Complete the routing table for B., A C E Cost VIA A 0 9 6 | ------------ ---------- B 5 11 7 0 | B ------------ ---------- C 2 0 3 | ------------ ---------- D 8 5 2 | ------------ ---------- E 6 8 0 | ------------ ---------- F 11 2 7 | ------------ ---------- Computer Science 360 Quiz O Name___________________ 1. The problem of counting to infinity is an affliction affecting: a. Distance vector protocols b. Shortest path first protocols c. Flooding d. Both a. and b. 2. The BGP protocol is use for routing a. within a single AS b. from AS to AS c. both 3. The BGP protocol is best characterized as a routing protocol of type a. flooding b. shortest path first c. distance vector 4. A BGP routing advertisement carries a list a. of IP addresses of routers b. of ASIDs of the autonomous through which the route systems through which the passes route passes. c. of routers within a single autonomous system 5. An IXP is best characterized as a. geographically distibuted b. a facility housing multiple collection of routers belonging routers of a single ISP to a single ISP c. geographically distibuted d. a facility housing multiple collection of routers belonging routers belonging to multiple to a multiple ISP ISP's