From devarshi at ieee.org Sat May 1 14:41:54 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Sat May 1 14:43:47 2004 Subject: [Csci551-talk] project B sample outputs Message-ID: <1083447714.25850.72.camel@maverick> Hi, I was just wondering when the project B sample outputs would be posted. --Devarshi From liyuan at pollux.usc.edu Sat May 1 15:55:05 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sat May 1 15:56:40 2004 Subject: [Csci551-talk] Please turn in timer library for proj B In-Reply-To: <005901c42f3c$272a66d0$6500a8c0@coolrocks> Message-ID: Dear Class, For students who use timer libarary provided by us, please tar all the files (including subdirectories) and turn in the tar ball. The command: tar cvf projb.tar projb Points will be deducted if we need to copy timer library for you. Thanks Yuan From liyuan at pollux.usc.edu Sat May 1 15:53:07 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sat May 1 15:57:41 2004 Subject: [Csci551-talk] Project B submition question In-Reply-To: <005901c42f3c$272a66d0$6500a8c0@coolrocks> Message-ID: Dear Class, For students who use timer libarary provided by us, please tar all the files (including subdirectories) and turn in the tar ball. The command: tar cvf projb.tar projb Thanks Yuan On Fri, 30 Apr 2004, Coolrocks wrote: > Hi, > I am using the provided solution code for project A. When submitting project B, how do I submit the timer code? The makefile in the provided solution goes to the "timer directory" for compilation of the timer code, but "submit" command doesn't support submitting whole directories. So shall I merge the timer code into the same directory as the other files and make changes to the "makefile" or else? > > Thanks! From liyuan at pollux.usc.edu Sat May 1 16:04:12 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sat May 1 16:05:41 2004 Subject: [Csci551-talk] Office hour change Message-ID: Dear Class, In order to accommodate those who can not show up on my Monday office hour, I will change my office hours this week to 7:00pm - 8:30pm (Tuesday) May 4. Please come to see me if you still have questions regarding project A grade. Thanks Yuan From johnh at ISI.EDU Sat May 1 17:25:45 2004 From: johnh at ISI.EDU (John Heidemann) Date: Sat May 1 17:27:41 2004 Subject: [Csci551-talk] Packet count In-Reply-To: <00ad01c42e62$362e8e70$6500a8c0@coolrocks> Message-ID: <200405020025.i420Pjne006030@dash.isi.edu> On Thu, 29 Apr 2004 20:21:17 PDT, "Coolrocks" wrote: >Hi, > Went through the updated specs, still have a doubt: Does a packet that is >originated and received by the same router also count as a "received packet"? > Or only a packet that is originated by some other router count as a >"received packet"? > >Thanks > > In general, if it's not specified you may chose either interpretation. If the interpretions can produce different output, indicate your choice in the README file. -John Heidemann From johnh at ISI.EDU Sat May 1 17:30:13 2004 From: johnh at ISI.EDU (John Heidemann) Date: Sat May 1 17:30:41 2004 Subject: [Csci551-talk] Question on Load Sensitive Routing Example In-Reply-To: <00c701c42daf$0a5229a0$7a503e9f@LATUALP> Message-ID: <200405020030.i420UD74006138@dash.isi.edu> On Wed, 28 Apr 2004 22:58:48 PDT, "Aaron Tu" wrote: >For the updated example given in section 4: > >S 5 15.0.0.1 11.0.0.1 >S 5 15.0.0.1 11.0.0.1 >S 5 15.0.0.1 11.0.0.1 ># after three packets, router 1 should up its MED value ># and propagate this to other nodes > >Since router 3 and 5 forward the packet, should they up their MED values >as well? Yes. >Should we wait until the packet reaches its destination before >propagating LSA message? >Or should we start immediately? >In this case, >should router 5 initiate the LSA message while the packet is still >trying to reach its destination? I think it would be much easier to what >until the packet reaches its destination before propagating LSA message. The assignment does not specify that level of timing detail. For things that are not specified you can chose. (Although I'm not sure why it would be easier to wait... adding synchronization generally takes extra effort.) -John Heidemann From johnh at ISI.EDU Sat May 1 17:33:15 2004 From: johnh at ISI.EDU (John Heidemann) Date: Sat May 1 17:34:41 2004 Subject: [Csci551-talk] bug fix for sample code input file handling In-Reply-To: <200404301535.i3UFZbQS000661@dash.isi.edu> Message-ID: <200405020033.i420XF6K006179@dash.isi.edu> On Fri, 30 Apr 2004 08:35:36 PDT, John Heidemann wrote: >A student previously posted on the list that the sample code's >handling of input files didn't match the project specification---the >sample code reads from an input file explicitly, not from stdin. > >In my initial response I suggested that the code handled either case, >but this statement was incorrect. > >A patch and/or replacement to main.cc should fix this problem, >allowing the sample code to read from stdin by default if no >command-line arguments are given. Since this IS how we will run tests >on the program, you should apply this patch to your code if you're >using the sample code in your project B. > >(We will NOT read inputs from config.conf when testing Proj B code!) Fix a bug... introduce a bug. :-( Although I intended Proj B to use the same config file input method as Proj A, a cut-and-paste error in Proj B DID define manager.conf (the method we'd used in prior years). We will therefore accept either of these two input methods. (I don't want to make things unnecessarily harder by changing the input method in the spec.) We will assume the default method is manager.conf (since that's in the spec). In either case, please define which method you chose in your README file. -John Heidemann From mqin at usc.edu Sat May 1 21:42:59 2004 From: mqin at usc.edu (min) Date: Sat May 1 21:45:42 2004 Subject: [Csci551-talk] Question on Load Sensitive Routing Example References: <200405020030.i420UD74006138@dash.isi.edu> Message-ID: <003001c42fff$fb6c20f0$6e547944@jimmy> according to our new handout, I think only router 3 and router 1 need up their MED value. Since router 5 originates the packet, router 3 forwards the packet, and router 1 recvs the packet. according to handout, routers forward and receive should update MED Am I right? Cheers Min ----- Original Message ----- From: "John Heidemann" To: "Aaron Tu" Cc: Sent: 2004?5?1? 17:30 Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > On Wed, 28 Apr 2004 22:58:48 PDT, "Aaron Tu" wrote: > >For the updated example given in section 4: > > > >S 5 15.0.0.1 11.0.0.1 > >S 5 15.0.0.1 11.0.0.1 > >S 5 15.0.0.1 11.0.0.1 > ># after three packets, router 1 should up its MED value > ># and propagate this to other nodes > > > >Since router 3 and 5 forward the packet, should they up their MED values > >as well? > > Yes. > > >Should we wait until the packet reaches its destination before > >propagating LSA message? > >Or should we start immediately? > >In this case, > >should router 5 initiate the LSA message while the packet is still > >trying to reach its destination? I think it would be much easier to what > >until the packet reaches its destination before propagating LSA message. > > The assignment does not specify that level of timing detail. For > things that are not specified you can chose. > > (Although I'm not sure why it would be easier to wait... adding > synchronization generally takes extra effort.) > > -John Heidemann > From coolrocks at tom.com Sat May 1 23:06:31 2004 From: coolrocks at tom.com (=?gb2312?B?QnJpYW4=?=) Date: Sat May 1 23:07:52 2004 Subject: =?gb2312?B?Rnc6IFtDc2NpNTUxLXRhbGtdIFF1ZXN0aW9uIG9uIExvYWQgU2Vuc2k=?= =?gb2312?B?dGl2ZSBSb3V0aW5nIEV4YW1wbGU=?= Message-ID: <40948FE7.000058.20950@bjapp5> I have the same doubt as Min: Shouldn't router 5 be considered as an originator, not a forwarder? Otherwise printing out "data-packets-originated" in stage 6 will be meaningless, i.e. there will be no "data packets originated" but only ones that are forwarded. Unless "forwared data packets" in stage 6 has a different "definition" in stage 7?? Please clarify! Thanks ----- Original Message ----- From: "min" To: "John Heidemann" ; "Aaron Tu" Cc: Sent: Saturday, May 01, 2004 9:42 PM Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > according to our new handout, I think only router 3 and router 1 need up > their MED value. Since router 5 originates the packet, router 3 forwards the > packet, and router 1 recvs the packet. > > according to handout, routers forward and receive should update MED > > Am I right? > > Cheers > Min > ----- Original Message ----- > From: "John Heidemann" > To: "Aaron Tu" > Cc: > Sent: 2004?5?1? 17:30 > Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > > > > On Wed, 28 Apr 2004 22:58:48 PDT, "Aaron Tu" wrote: > > >For the updated example given in section 4: > > > > > >S 5 15.0.0.1 11.0.0.1 > > >S 5 15.0.0.1 11.0.0.1 > > >S 5 15.0.0.1 11.0.0.1 > > ># after three packets, router 1 should up its MED value > > ># and propagate this to other nodes > > > > > >Since router 3 and 5 forward the packet, should they up their MED values > > >as well? > > > > Yes. > > > > >Should we wait until the packet reaches its destination before > > >propagating LSA message? > > >Or should we start immediately? > > >In this case, > > >should router 5 initiate the LSA message while the packet is still > > >trying to reach its destination? I think it would be much easier to what > > >until the packet reaches its destination before propagating LSA message. > > > > The assignment does not specify that level of timing detail. For > > things that are not specified you can chose. > > > > (Although I'm not sure why it would be easier to wait... adding > > synchronization generally takes extra effort.) > > > > -John Heidemann > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.isi.edu/pipermail/csci551-talk/attachments/20040502/1107c571/attachment-0001.html From coolrocks at tom.com Sat May 1 23:31:35 2004 From: coolrocks at tom.com (Coolrocks) Date: Sat May 1 23:45:49 2004 Subject: Fw: [Csci551-talk] Question on Load Sensitive Routing Example Message-ID: <003a01c43010$d485e050$6500a8c0@coolrocks> Oops, the first message turned out to be unexpectedly messy, let me try again:) I have the same doubt as Min: Shouldn't router 5 be considered as an originator, not a forwarder? Otherwise printing out "data-packets-originated" in stage 6 will be meaningless, i.e. there will be no "data packets originated" but only ones that are forwarded. Unless "forwarded data packets" in stage 6 has a different "definition" in stage 7?? Please clarify! Thanks ----- Original Message ----- From: "min" To: "John Heidemann" ; "Aaron Tu" Cc: Sent: Saturday, May 01, 2004 9:42 PM Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > according to our new handout, I think only router 3 and router 1 need up > their MED value. Since router 5 originates the packet, router 3 forwards the > packet, and router 1 recvs the packet. > > according to handout, routers forward and receive should update MED > > Am I right? > > Cheers > Min > ----- Original Message ----- > From: "John Heidemann" > To: "Aaron Tu" > Cc: > Sent: 2004?5?1? 17:30 > Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > > > > On Wed, 28 Apr 2004 22:58:48 PDT, "Aaron Tu" wrote: > > >For the updated example given in section 4: > > > > > >S 5 15.0.0.1 11.0.0.1 > > >S 5 15.0.0.1 11.0.0.1 > > >S 5 15.0.0.1 11.0.0.1 > > ># after three packets, router 1 should up its MED value > > ># and propagate this to other nodes > > > > > >Since router 3 and 5 forward the packet, should they up their MED values > > >as well? > > > > Yes. > > > > >Should we wait until the packet reaches its destination before > > >propagating LSA message? > > >Or should we start immediately? > > >In this case, > > >should router 5 initiate the LSA message while the packet is still > > >trying to reach its destination? I think it would be much easier to what > > >until the packet reaches its destination before propagating LSA message. > > > > The assignment does not specify that level of timing detail. For > > things that are not specified you can chose. > > > > (Although I'm not sure why it would be easier to wait... adding > > synchronization generally takes extra effort.) > > > > -John Heidemann > > > > From wpun at usc.edu Sun May 2 18:42:13 2004 From: wpun at usc.edu (Raymond Pun) Date: Sun May 2 18:43:43 2004 Subject: [Csci551-talk] unable to submit homework 4 Message-ID: <06b701c430af$dbfcbbc0$a701a8c0@raymond> Hi TA/Prof., I try to submit hw4 but failed with the following message: The tag "hw4" is not valid. You must specify one of the following: hw1 hw2 hw3 proja projb Could you please add the hw4 tag? Many thanks. regards, Raymond From liyuan at pollux.usc.edu Sun May 2 19:22:09 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 2 19:23:44 2004 Subject: [Csci551-talk] unable to submit homework 4 In-Reply-To: <06b701c430af$dbfcbbc0$a701a8c0@raymond> Message-ID: Problem should be fixed now. Please try it again. Thanks Yuan On Sun, 2 May 2004, Raymond Pun wrote: > Hi TA/Prof., > > I try to submit hw4 but failed with the following message: > > The tag "hw4" is not valid. You must specify one of the following: > hw1 hw2 hw3 proja projb > > Could you please add the hw4 tag? > > Many thanks. > > regards, > Raymond > From liyuan at pollux.usc.edu Sun May 2 19:40:33 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 2 19:41:44 2004 Subject: [Csci551-talk] sample input and output for project B is ready Message-ID: Pleae check them at : http://www-scf.usc.edu/~yuanl/551/projb.html Please notice that the output is not "guaranteed". Thanks Yuan From johnh at ISI.EDU Mon May 3 08:14:13 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 3 08:15:46 2004 Subject: [Csci551-talk] homework 3 key posted to class web site Message-ID: <200405031514.i43FEDwZ018302@dash.isi.edu> The homework 3 key has been posted to class web site. -John Heidemann From sblackmo at usc.edu Mon May 3 17:48:57 2004 From: sblackmo at usc.edu (stephen blackmon) Date: Mon May 3 17:50:10 2004 Subject: [Csci551-talk] output expectations Message-ID: <1344306c2d1a.40968609@usc.edu> Could you please comment on what you will be looking for in our output files? Will you expect a straight 'diff' to return no comments? I find my number of LSA Route Tables are different and sometimes duplicated, but the routing works as expected. Should I go out of my way to make the output match exactly (not sure if i even can) or just ensure the requirements of the handout are fulfilled? Thanks. From liyuan at pollux.usc.edu Mon May 3 20:40:16 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Mon May 3 20:41:16 2004 Subject: [Csci551-talk] output expectations In-Reply-To: <1344306c2d1a.40968609@usc.edu> Message-ID: No, we do not expect your ouput files exactly match the sample outputs. Like project A, you only need to make sure that a copy of the routing table is printed whenever it changes. And please do not print out routing table too frequently. Programs which print out unreasonablely large size of routing tables (like 1 thousand lines) will get points deducted for Project B. Thanks Yuan On Mon, 3 May 2004, stephen blackmon wrote: > Could you please comment on what you will be looking for in our output files? Will you expect a straight 'diff' to return no comments? I find my number of LSA Route Tables are different and sometimes duplicated, but the routing works as expected. Should I go out of my way to make the output match exactly (not sure if i even can) or just ensure the requirements of the handout are fulfilled? > > Thanks. > From johnh at ISI.EDU Wed May 5 19:57:54 2004 From: johnh at ISI.EDU (John Heidemann) Date: Wed May 5 19:58:18 2004 Subject: [Csci551-talk] Question on Load Sensitive Routing Example In-Reply-To: <003001c42fff$fb6c20f0$6e547944@jimmy> Message-ID: <200405060257.i462vsPP009527@dash.isi.edu> On Sat, 01 May 2004 21:42:59 PDT, "min" wrote: >according to our new handout, I think only router 3 and router 1 need up >their MED value. Since router 5 originates the packet, router 3 forwards the >packet, and router 1 recvs the packet. > >according to handout, routers forward and receive should update MED > >Am I right? Yes, you're right that routers that originate the packet are not considered to have forwarded the packet for MED purposes. (My simple yes below needs qualification.) -John Heidemann > >Cheers >Min >----- Original Message ----- >From: "John Heidemann" >To: "Aaron Tu" >Cc: >Sent: 2004?5?1? 17:30 >Subject: Re: [Csci551-talk] Question on Load Sensitive Routing Example > > >> On Wed, 28 Apr 2004 22:58:48 PDT, "Aaron Tu" wrote: >> >For the updated example given in section 4: >> > >> >S 5 15.0.0.1 11.0.0.1 >> >S 5 15.0.0.1 11.0.0.1 >> >S 5 15.0.0.1 11.0.0.1 >> ># after three packets, router 1 should up its MED value >> ># and propagate this to other nodes >> > >> >Since router 3 and 5 forward the packet, should they up their MED values >> >as well? >> >> Yes. >> >> >Should we wait until the packet reaches its destination before >> >propagating LSA message? >> >Or should we start immediately? >> >In this case, >> >should router 5 initiate the LSA message while the packet is still >> >trying to reach its destination? I think it would be much easier to what >> >until the packet reaches its destination before propagating LSA message. >> >> The assignment does not specify that level of timing detail. For >> things that are not specified you can chose. >> >> (Although I'm not sure why it would be easier to wait... adding >> synchronization generally takes extra effort.) >> >> -John Heidemann >> From mqin at usc.edu Thu May 6 12:11:01 2004 From: mqin at usc.edu (Min Qin) Date: Thu May 6 12:13:21 2004 Subject: [Csci551-talk] problem about oscillation and non-oscillation Message-ID: <000e01c4339d$decf0dd0$51477d80@mqin> Dear all, the oscillation part seems wierd now. For handout stage 8, i think the ip address 11.0.0.0/8 and 12.0.0.8/8 should be modified to 11.0.0.0/8 and 11.0.0.8/8. This will make the route oscillate between 5-3-1 and 5-4-2 However, this makes the non-oscillation part hard. How to make the route non-occillate indefinitely? as long as router 1 and router 2 has the same ip prefx/mask, there will definitely exist osscillation. Even a large initial Med value can only guarantee the route not to oscillate in a number of packets, Also, our input does not provide initial Med value for internal routers(they are equal to 0). of course, sending to router 5 will not cause occilation, but is this what the part requires? Cheers Min Qin -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.isi.edu/pipermail/csci551-talk/attachments/20040506/4c69ab80/attachment.html From johnh at ISI.EDU Thu May 6 16:15:00 2004 From: johnh at ISI.EDU (John Heidemann) Date: Thu May 6 16:16:30 2004 Subject: [Csci551-talk] office hours Message-ID: <200405062315.i46NF0Yl021918@dash.isi.edu> Several students have asked about office hours over finals week. I WILL have office hours this Friday afternoon as usual. (This will be my last formal office hours for the summer.) -John Heidemann From johnh at ISI.EDU Fri May 7 10:43:50 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri May 7 10:44:43 2004 Subject: [Csci551-talk] problem about oscillation and non-oscillation In-Reply-To: <000e01c4339d$decf0dd0$51477d80@mqin> Message-ID: <200405071743.i47HhobR032638@dash.isi.edu> On Thu, 06 May 2004 12:11:01 PDT, Min Qin wrote: >Dear all, > the oscillation part seems wierd now. For handout stage 8, i think the ip >address 11.0.0.0/8 and 12.0.0.8/8 should be modified to 11.0.0.0/8 and >11.0.0.8/8. This will make the route oscillate between 5-3-1 and 5-4-2 > However, this makes the non-oscillation part hard. How to make the route >non-occillate indefinitely? as long as router 1 and router 2 has the same ip >prefx/mask, there will definitely exist osscillation. Even a large initial Med >value can only guarantee the route not to oscillate in a number of packets, >Also, our input does not provide initial Med value for internal routers(they >are equal to 0). > of course, sending to router 5 will not cause occilation, but is this what >the part requires? > >Cheers >Min Qin > > If you don't believe you can generate a case that does not exhibit oscillation or non-oscillation, please state that fact and WHY in the your README file. Then describe (in the README) minimum modification to the assignment you could make that would allow you to demonstrate oscillation or non-oscillation. -John Heidemann From atu011 at earthlink.net Fri May 7 13:49:05 2004 From: atu011 at earthlink.net (Aaron Tu) Date: Fri May 7 13:50:25 2004 Subject: [Csci551-talk] osc/noosc cases In-Reply-To: <200405071900.i47J0Ei16463@gamma.isi.edu> Message-ID: <000001c43474$bd423520$7a503e9f@LATUALP> The assignment said we can demonstrate osc/noosc with 8 router commands. Do these commands include the 'w' commands? For the osc case, is it sufficient to show that it switches to another route (once) or do we have to show that it switches back to the original route (thought this can be inferred) to be considered oscillation? Similar question for the noosc case. Aaron From pilani at usc.edu Fri May 7 23:48:15 2004 From: pilani at usc.edu (rahul pilani) Date: Fri May 7 23:48:29 2004 Subject: [Csci551-talk] Config File for Router with multiple responsible address spaces Message-ID: <8491a8bf5929.409c203f@usc.edu> Hi, I was going through my configuration file and noticed something. For Project B.. a router can be responsible for multiple routes right? So how will this be done in the configuration file??.. it seems that we are currently parsing directly.. with 1st route going to router 1, 2nd going to router 2 etc. Will this change the Config File fomat also?.. below is an example of the current way, responsible routes are being assidned. Please indicate if this way is still valid for Project B. ..... # addresses and masks for each internal router, in order # in the form a.b.c.d e where e is the prefix length. # last entry is 0.0.0.0 0 # The first line is for router 1, 2nd for router 2, etc. 10.1.0.0 16 10.2.0.0 16 192.168.123.0 24 192.168.101.0 24 192.168.200.128 25 0.0.0.0 0 # external routes needed for policy routing ... ... Thanks, Rahul Pilani From payalsin at usc.edu Sat May 8 20:03:35 2004 From: payalsin at usc.edu (payal singh) Date: Sat May 8 20:04:26 2004 Subject: [Csci551-talk] Slip day Message-ID: hi, wanted to request for a slip day...do i mail the TA.. thanks.. payal From liyuan at pollux.usc.edu Sat May 8 23:07:23 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sat May 8 23:08:29 2004 Subject: [Csci551-talk] Slip day In-Reply-To: Message-ID: Please follow the same procedure as Project A. Please check details in my previous posts on the mail list. http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000762.html http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000763.html Thanks Yuan On Sat, 8 May 2004, payal singh wrote: > hi, > > wanted to request for a slip day...do i mail the TA.. > > thanks.. > > payal > From wpun at usc.edu Sat May 8 23:33:44 2004 From: wpun at usc.edu (Raymond Pun) Date: Sat May 8 23:22:39 2004 Subject: [Csci551-talk] Slip day References: Message-ID: <032d01c4358f$93d6f470$a701a8c0@raymond> Hi, So, i should send an email to liyuan@pollux.usc.edu with subject line: CSCI551 ProjB Slip Day Request, right? Will there be any confirmation reply? many thanks. regards, Raymond ----- Original Message ----- From: "Yuan Li" To: "payal singh" Cc: Sent: Saturday, May 08, 2004 23:07 Subject: Re: [Csci551-talk] Slip day > Please follow the same procedure as Project A. > > Please check details in my previous posts on the mail list. > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000762.html > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000763.html > > Thanks > > Yuan > > > On Sat, 8 May 2004, payal singh wrote: > > > hi, > > > > wanted to request for a slip day...do i mail the TA.. > > > > thanks.. > > > > payal > > > > From liyuan at pollux.usc.edu Sun May 9 00:00:40 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 9 00:01:37 2004 Subject: [Csci551-talk] Slip day In-Reply-To: <032d01c4358f$93d6f470$a701a8c0@raymond> Message-ID: yes, that is correct. I will send out the confirmation emails right after the submission deadline. And I will also post a message on the mail list. Please check at that time. Thanks Yuan On Sat, 8 May 2004, Raymond Pun wrote: > Hi, > > So, i should send an email to liyuan@pollux.usc.edu with subject line: > > CSCI551 ProjB Slip Day Request, right? Will there be any confirmation > reply? > > many thanks. > > regards, > Raymond > > ----- Original Message ----- > From: "Yuan Li" > To: "payal singh" > Cc: > Sent: Saturday, May 08, 2004 23:07 > Subject: Re: [Csci551-talk] Slip day > > > > Please follow the same procedure as Project A. > > > > Please check details in my previous posts on the mail list. > > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000762.html > > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000763.html > > > > Thanks > > > > Yuan > > > > > > On Sat, 8 May 2004, payal singh wrote: > > > > > hi, > > > > > > wanted to request for a slip day...do i mail the TA.. > > > > > > thanks.. > > > > > > payal > > > > > > > > From liyuan at pollux.usc.edu Sun May 9 00:02:23 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 9 00:03:26 2004 Subject: [Csci551-talk] Project B slip day request procedure Message-ID: Send an email to liyuan at pollux.usc.edu with the following subject line: CSCI551 ProjB Slip Day Request You do not need to put any text in the body. And I will ignore all the requests that does not follow this guideline. Thanks Yuan Li From wpun at usc.edu Sun May 9 00:58:51 2004 From: wpun at usc.edu (Raymond Pun) Date: Sun May 9 00:47:26 2004 Subject: [Csci551-talk] Slip day References: Message-ID: <037101c4359b$78366410$a701a8c0@raymond> Hi, But then..... how do we know our request is succeesfully recevied? If after the deadline I found that my reqeust is not received.... what should I do? Can we have the confirmation before the deadline? many thanks. regards, Raymond ----- Original Message ----- From: "Yuan Li" To: "Raymond Pun" Cc: Sent: Sunday, May 09, 2004 00:00 Subject: Re: [Csci551-talk] Slip day > yes, that is correct. > > I will send out the confirmation emails right after the submission > deadline. And I will also post a message on the mail list. Please check at > that time. > > Thanks > > Yuan > > > On Sat, 8 May 2004, Raymond Pun wrote: > > > Hi, > > > > So, i should send an email to liyuan@pollux.usc.edu with subject line: > > > > CSCI551 ProjB Slip Day Request, right? Will there be any confirmation > > reply? > > > > many thanks. > > > > regards, > > Raymond > > > > ----- Original Message ----- > > From: "Yuan Li" > > To: "payal singh" > > Cc: > > Sent: Saturday, May 08, 2004 23:07 > > Subject: Re: [Csci551-talk] Slip day > > > > > > > Please follow the same procedure as Project A. > > > > > > Please check details in my previous posts on the mail list. > > > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000762.html > > > http://mailman.isi.edu/pipermail/csci551-talk/2004-March/000763.html > > > > > > Thanks > > > > > > Yuan > > > > > > > > > On Sat, 8 May 2004, payal singh wrote: > > > > > > > hi, > > > > > > > > wanted to request for a slip day...do i mail the TA.. > > > > > > > > thanks.. > > > > > > > > payal > > > > > > > > > > > > > > From rshroff at usc.edu Sun May 9 09:00:08 2004 From: rshroff at usc.edu (rajesh shroff) Date: Sun May 9 09:01:27 2004 Subject: [Csci551-talk] stage 6 & 7 Message-ID: hello, Just a quick check. 1. MED increment ONLY has to happen in stage 7. Right? 2. In stage 6 even though the forwarded/received packets exceed 3, we do not have to change MED value. Right? Please let me know about these two statements. Regards Rajesh From atu011 at earthlink.net Sun May 9 11:34:50 2004 From: atu011 at earthlink.net (Aaron Tu) Date: Sun May 9 11:35:26 2004 Subject: [Csci551-talk] output In-Reply-To: <200405081900.i48J04i04475@gamma.isi.edu> Message-ID: <000001c435f4$50751800$6701a8c0@LATUALP> I am using the sample Project A source code. Are the sample outputs generated from the sample Project A? The final stage of the routing table is the same for each sample output; however, the order of the entries is not the same. Also, I have more intermediate steps in all of my output files as compared with the sample output files. Is this a concern, since the generation of the routing table is a function of Project A? Aaron From stambe at usc.edu Sun May 9 11:38:26 2004 From: stambe at usc.edu (swara tambe) Date: Sun May 9 11:39:27 2004 Subject: [Csci551-talk] stage 6 & 7 Message-ID: <1630d0144cab.409e1832@usc.edu> Hi Rajesh, I believe the MED changes are only to be made in stage 7. Stage 6 is devised only to test whether you can forward packets correctly. -Swara ----- Original Message ----- From: rajesh shroff Date: Sunday, May 9, 2004 9:00 am Subject: [Csci551-talk] stage 6 & 7 > hello, > Just a quick check. > > 1. MED increment ONLY has to happen in stage 7. Right? > 2. In stage 6 even though the forwarded/received packets exceed 3, we do > not have to change MED value. Right? > > Please let me know about these two statements. > > Regards > Rajesh > > From stambe at usc.edu Sun May 9 11:52:56 2004 From: stambe at usc.edu (swara tambe) Date: Sun May 9 11:54:27 2004 Subject: [Csci551-talk] Stage 6 Message-ID: <373df8d11d31.409e1b98@usc.edu> Hi all, For stage 6, if a router receives a destination IP (from the manager) which is part of the router's own address space, does the router still have to print "Sending data" and "Received data", or does it just not print anything at all? Thanks, Swara From liyuan at pollux.usc.edu Sun May 9 12:03:52 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 9 12:04:27 2004 Subject: [Csci551-talk] output In-Reply-To: <000001c435f4$50751800$6701a8c0@LATUALP> Message-ID: You do not need to make your output look exactly the same as the sample outputs. Yuan On Sun, 9 May 2004, Aaron Tu wrote: > I am using the sample Project A source code. Are the sample outputs > generated from the sample Project A? The final stage of the routing > table is the same for each sample output; however, the order of the > entries is not the same. Also, I have more intermediate steps in all of > my output files as compared with the sample output files. Is this a > concern, since the generation of the routing table is a function of > Project A? > > Aaron > From liyuan at pollux.usc.edu Sun May 9 12:19:49 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 9 12:21:26 2004 Subject: [Csci551-talk] Stage 6 In-Reply-To: <373df8d11d31.409e1b98@usc.edu> Message-ID: There is such a case in the sample testcase for stage 6. And please print out the "sending" "receiving" messages and please also take those kinds of packets into consideration when couting # of messages sent and received. Yuan On Sun, 9 May 2004, swara tambe wrote: > Hi all, > > For stage 6, if a router receives a destination IP (from the manager) which is part of the router's own address space, does the router still have to print "Sending data" and "Received data", or does it just not print anything at all? > > Thanks, > Swara > From liyuan at pollux.usc.edu Sun May 9 12:21:36 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Sun May 9 12:22:26 2004 Subject: [Csci551-talk] stage 6 & 7 In-Reply-To: <1630d0144cab.409e1832@usc.edu> Message-ID: Correct. You do not need to change MED value in stage6. Yuan On Sun, 9 May 2004, swara tambe wrote: > Hi Rajesh, > > I believe the MED changes are only to be made in stage 7. Stage 6 is devised only to test whether you can forward packets correctly. > > -Swara > > > ----- Original Message ----- > From: rajesh shroff > Date: Sunday, May 9, 2004 9:00 am > Subject: [Csci551-talk] stage 6 & 7 > > > hello, > > Just a quick check. > > > > 1. MED increment ONLY has to happen in stage 7. Right? > > 2. In stage 6 even though the forwarded/received packets exceed 3, we do > > not have to change MED value. Right? > > > > Please let me know about these two statements. > > > > Regards > > Rajesh > > > > > From gvagarwa at usc.edu Sun May 9 17:19:07 2004 From: gvagarwa at usc.edu (gaurav agarwal) Date: Sun May 9 17:20:28 2004 Subject: [Csci551-talk] When to increase MED? Message-ID: <2231112c5f5b.409e680b@usc.edu> Hi Do we increase MED when data forwarded or recieved is 3 OR sum of data forwarded and received is 3 Also what do you exactly mean by local net? Thanks Gaurav From rshroff at usc.edu Sun May 9 17:42:20 2004 From: rshroff at usc.edu (rajesh shroff) Date: Sun May 9 17:43:28 2004 Subject: [Csci551-talk] stage 8 Message-ID: <36cfaf343f2e.409e6d7c@usc.edu> Hi, for stage 7 it is assumed that load sensitive routing cannot proceed unless multiple routers manage the same address space. But for stage 8, we are supposed to do load sensitive routing without routers managing the same space (according to the inputfile given for stage 8 ).. I tried many configurations, but they dont seem to work. My question is,, whether we are allowed to change the router managed address spaces (many routers managing one address space), to make load sensitive routing feasible. Thanks, From wpun at usc.edu Sun May 9 18:29:51 2004 From: wpun at usc.edu (Raymond Pun) Date: Sun May 9 18:18:30 2004 Subject: [Csci551-talk] stage 8 References: <36cfaf343f2e.409e6d7c@usc.edu> Message-ID: <050a01c4362e$4b651dd0$a701a8c0@raymond> Hi, I think we cannot make load sensitive routing without havin multiple routers manage the same prefix. MED is related to a prefix. If there is only 1 router manage that prefix, then, there is only 1 route to that prefix as there doesn't exists another MED value to compare. So, I guess the line "12.0.0.0 8" should be "11.0.0.0 8" just like the one as stage 7. Please correct me if I am wrong. Many thanks. regards, Raymond ----- Original Message ----- From: "rajesh shroff" To: Sent: Sunday, May 09, 2004 17:42 Subject: [Csci551-talk] stage 8 > Hi, > for stage 7 it is assumed that load sensitive routing cannot proceed > unless multiple routers manage the same address space. But for stage 8, > we are supposed to do load sensitive routing without routers managing > the same space (according to the inputfile given for stage 8 ).. I > tried many configurations, but they dont seem to work. My question is,, > whether we are allowed to change the router managed address spaces (many > routers managing one address space), to make load sensitive routing > feasible. > > Thanks, > > > > > From wpun at usc.edu Sun May 9 23:58:51 2004 From: wpun at usc.edu (Raymond Pun) Date: Sun May 9 23:47:33 2004 Subject: [Csci551-talk] question about s7-1.in Message-ID: <055501c4365c$40fcffb0$a701a8c0@raymond> Hi, I have a question. from s7-1.in we have router --> prefix 1 --> 11.0.0.0/8 2 --> 11.0.0.0/8 3 --> 13.0.0.0/8 4 --> 14.0.0.0/8 5 --> 15.0.0.0/8 So, both routers 1 and 2 got the prefix 11.0.0.0/8 After 1 received 3 data packets, its MED goes to 1. The quetions is... If router 3 is going to send packet to 11.0.0.1. It destinate the packet at router 2. But the packet will still go to router 1 as next hop. When router 1 got this packet, will it forwards it to router 2? >From s7-1.out, we have "2 11.0.0.0/8 2 1 (med=0)". However, router 1 also have the prefix 11.0.0.0/8. It should consume the packet logically. So... should router 1 still forward it? or consume it? Thanks. regards, Raymond From jayant.v.rao at usc.edu Mon May 10 00:16:54 2004 From: jayant.v.rao at usc.edu (Jayant V. Rao) Date: Mon May 10 00:16:27 2004 Subject: [Csci551-talk] question about s7-1.in In-Reply-To: <055501c4365c$40fcffb0$a701a8c0@raymond> Message-ID: <200405100714.i4A7Ewk00149@vapor.isi.edu> I have a related question - Topo: 1-2, 1-3, 2-4, 3-4 1, 2 manage 11.0.0.0/8. Suppose now 1 has med=1, and 2 has med=0. 3 sends a message to 11.0.0.1. it should go to 2, and avoid 1 right? The correct path is: 3-4-2 Or is it ok to send 3-1-2? Because this is what seems to be happening below, sending to 2 thru 1, even though we should avoid 1. Clarifications please, Jayant -----Original Message----- From: csci551-talk-bounces@mailman.isi.edu [mailto:csci551-talk-bounces@mailman.isi.edu] On Behalf Of Raymond Pun Sent: Sunday, May 09, 2004 11:59 PM To: csci551-talk@ISI.EDU Subject: [Csci551-talk] question about s7-1.in Hi, I have a question. from s7-1.in we have router --> prefix 1 --> 11.0.0.0/8 2 --> 11.0.0.0/8 3 --> 13.0.0.0/8 4 --> 14.0.0.0/8 5 --> 15.0.0.0/8 So, both routers 1 and 2 got the prefix 11.0.0.0/8 After 1 received 3 data packets, its MED goes to 1. The quetions is... If router 3 is going to send packet to 11.0.0.1. It destinate the packet at router 2. But the packet will still go to router 1 as next hop. When router 1 got this packet, will it forwards it to router 2? >From s7-1.out, we have "2 11.0.0.0/8 2 1 (med=0)". However, router 1 also have the prefix 11.0.0.0/8. It should consume the packet logically. So... should router 1 still forward it? or consume it? Thanks. regards, Raymond From wpun at usc.edu Mon May 10 01:19:02 2004 From: wpun at usc.edu (Raymond Pun) Date: Mon May 10 01:07:27 2004 Subject: [Csci551-talk] manager.conf? Message-ID: <057501c43667$74771870$a701a8c0@raymond> Hi, I suddenly dicovered that we should read a file named manager.conf instead of having stdin. So... we can hardcode the program to read manager.conf instead of reading input from the stdin, right? thanks regards, Raymond From wpun at usc.edu Mon May 10 01:40:14 2004 From: wpun at usc.edu (Raymond Pun) Date: Mon May 10 01:28:27 2004 Subject: [Csci551-talk] about stage 8 & 9 input/output Message-ID: <05a901c4366a$6a96fd90$a701a8c0@raymond> Hi, For stage 8, should the outfiles be s8-X.out or still s7-X.out? For the non-oscillation input, should be 1st line be stage 8 or stage 9? And should be output be s7X.out, s8-X.out or s9-X.out? please clarify. Many thanks. regards, Raymond From raymondpun at graduate.hku.hk Sun May 9 23:56:27 2004 From: raymondpun at graduate.hku.hk (Raymond Pun) Date: Mon May 10 08:17:17 2004 Subject: [Csci551-talk] load sensitive route Message-ID: <054501c4365b$ead75180$a701a8c0@raymond> Hi, I have a question. from s7-1.in we have router --> prefix 1 --> 11.0.0.0/8 2 --> 11.0.0.0/8 3 --> 13.0.0.0/8 4 --> 14.0.0.0/8 5 --> 15.0.0.0/8 So, both routers 1 and 2 got the prefix 11.0.0.0/8 After 1 received 3 data packets, its MED goes to 1. The quetions is... If router 3 is going to send packet to 11.0.0.1. It destinate the packet at router 2. But the packet will still go to router 1 as next hop. When router 1 got this packet, will it forwards it to router 2? >From s7-1.out, we have "2 11.0.0.0/8 2 1 (med=0)". However, router 1 also have the prefix 11.0.0.0/8. It should consume the packet logically. So... should router 1 still forward it? or consume it? Thanks. regards, Raymond From liyuan at pollux.usc.edu Mon May 10 13:48:28 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Mon May 10 13:49:35 2004 Subject: [Csci551-talk] Slip Day confirmation Email Message-ID: I have sent out a confirmation email to students who requested a slip day for Project B. If you believe that you sent the request but did not receive a confirmation email, please check the request format I posted on the mail list and send me the request again ASAP. Thanks Yuan From johnh at ISI.EDU Mon May 10 16:41:20 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:42:33 2004 Subject: [Csci551-talk] stage 6 & 7 In-Reply-To: Message-ID: <200405102341.i4ANfK2h028951@dash.isi.edu> On Sun, 09 May 2004 12:21:36 PDT, Yuan Li wrote: > >Correct. You do not need to change MED value in stage6. But you do need to use the changing MEDs in stages 7 *and later*. -John Heidemann >On Sun, 9 May 2004, swara tambe wrote: > >> Hi Rajesh, >> >> I believe the MED changes are only to be made in stage 7. Stage 6 is devised only to test whether you can forward packets correctly. >> >> -Swara >> >> >> ----- Original Message ----- >> From: rajesh shroff >> Date: Sunday, May 9, 2004 9:00 am >> Subject: [Csci551-talk] stage 6 & 7 >> >> > hello, >> > Just a quick check. >> > >> > 1. MED increment ONLY has to happen in stage 7. Right? >> > 2. In stage 6 even though the forwarded/received packets exceed 3, we do >> > not have to change MED value. Right? >> > >> > Please let me know about these two statements. >> > >> > Regards >> > Rajesh >> > >> > >> From johnh at ISI.EDU Mon May 10 16:45:20 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:45:30 2004 Subject: [Csci551-talk] When to increase MED? In-Reply-To: <2231112c5f5b.409e680b@usc.edu> Message-ID: <200405102345.i4ANjK1F029000@dash.isi.edu> On Sun, 09 May 2004 17:19:07 PDT, gaurav agarwal wrote: >Hi > >Do we increase MED when >data forwarded or recieved is 3 >OR >sum of data forwarded and received is 3 > >Also what do you exactly mean by local net? > >Thanks >Gaurav I think the spec is clear on this point. -John Heidemann From johnh at ISI.EDU Mon May 10 16:47:59 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:49:40 2004 Subject: [Csci551-talk] stage 8 In-Reply-To: <36cfaf343f2e.409e6d7c@usc.edu> Message-ID: <200405102347.i4ANlxIf029019@dash.isi.edu> On Sun, 09 May 2004 17:42:20 PDT, rajesh shroff wrote: >Hi, >for stage 7 it is assumed that load sensitive routing cannot proceed >unless multiple routers manage the same address space. But for stage 8, >we are supposed to do load sensitive routing without routers managing >the same space (according to the inputfile given for stage 8 ).. I >tried many configurations, but they dont seem to work. My question is,, >whether we are allowed to change the router managed address spaces (many >routers managing one address space), to make load sensitive routing >feasible. > >Thanks, > > > We talked on the mailing list already about ways to demonstrate osciallation. Since you are providing the input file s8osc.in, you should make whatever changes to that input file to demonstrate osciallation. (The given sample input file is a sample---not the rule about what you should give!) -John Heidemann From johnh at ISI.EDU Mon May 10 16:54:28 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:55:40 2004 Subject: [Csci551-talk] about stage 8 & 9 input/output In-Reply-To: <05a901c4366a$6a96fd90$a701a8c0@raymond> Message-ID: <200405102354.i4ANsSQN029094@dash.isi.edu> On Mon, 10 May 2004 01:40:14 PDT, "Raymond Pun" wrote: >Hi, > >For stage 8, should the outfiles be s8-X.out or still s7-X.out? >For the non-oscillation input, should be 1st line be stage 8 or stage 9? And >should be output be s7X.out, s8-X.out or s9-X.out? > >please clarify. Many thanks. > >regards, >Raymond The output file numbers should match the stage. -John Heidemann From johnh at ISI.EDU Mon May 10 16:53:38 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:55:44 2004 Subject: [Csci551-talk] manager.conf? In-Reply-To: <057501c43667$74771870$a701a8c0@raymond> Message-ID: <200405102353.i4ANrcsu029075@dash.isi.edu> On Mon, 10 May 2004 01:19:02 PDT, "Raymond Pun" wrote: >Hi, > >I suddenly dicovered that we should read a file named manager.conf instead >of having stdin. >So... we can hardcode the program to read manager.conf instead of reading >input from the stdin, right? > >thanks > >regards, >Raymond This issue was discussed previously on the mailing list: you can use EITHER stdin OR manager.conf; please document which you use in README. -John Heidemann From johnh at ISI.EDU Mon May 10 16:56:15 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:56:31 2004 Subject: [Csci551-talk] osc/noosc cases In-Reply-To: <000001c43474$bd423520$7a503e9f@LATUALP> Message-ID: <200405102356.i4ANuFif029132@dash.isi.edu> On Fri, 07 May 2004 13:49:05 PDT, "Aaron Tu" wrote: >The assignment said we can demonstrate osc/noosc with 8 router commands. >Do these commands include the 'w' commands? For the osc case, is it >sufficient to show that it switches to another route (once) or do we >have to show that it switches back to the original route (thought this >can be inferred) to be considered oscillation? Similar question for the >noosc case. > >Aaron You need to do whatever you think is necessary to demonstrate oscillation. -John Heidemann From johnh at ISI.EDU Mon May 10 16:57:46 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 16:58:29 2004 Subject: [Csci551-talk] Config File for Router with multiple responsible address spaces In-Reply-To: <8491a8bf5929.409c203f@usc.edu> Message-ID: <200405102357.i4ANvkds029151@dash.isi.edu> On Fri, 07 May 2004 23:48:15 PDT, rahul pilani wrote: >Hi, >I was going through my configuration file and noticed something. >For Project B.. a router can be responsible for multiple routes right? > >So how will this be done in the configuration file??.. >it seems that we are currently parsing directly.. with 1st route going to router 1, 2nd going to router 2 etc. Will this change the Config File fomat also?.. >below is an example of the current way, responsible routes are being assidned. Please indicate if this way is still valid for Project B. > > >..... ># addresses and masks for each internal router, in order ># in the form a.b.c.d e where e is the prefix length. ># last entry is 0.0.0.0 0 ># The first line is for router 1, 2nd for router 2, etc. >10.1.0.0 16 >10.2.0.0 16 >192.168.123.0 24 >192.168.101.0 24 >192.168.200.128 25 >0.0.0.0 0 ># external routes needed for policy routing >... >... > >Thanks, >Rahul Pilani Having one router support multiple internal routes would require a config file change. I don't know of any discussion of that capability or changing the config file. -John Heidemann From johnh at ISI.EDU Mon May 10 17:00:52 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 17:02:31 2004 Subject: [Csci551-talk] question about s7-1.in In-Reply-To: <055501c4365c$40fcffb0$a701a8c0@raymond> Message-ID: <200405110000.i4B00qCp029276@dash.isi.edu> On Sun, 09 May 2004 23:58:51 PDT, "Raymond Pun" wrote: >Hi, > >I have a question. >from s7-1.in >we have >router --> prefix >1 --> 11.0.0.0/8 >2 --> 11.0.0.0/8 >3 --> 13.0.0.0/8 >4 --> 14.0.0.0/8 >5 --> 15.0.0.0/8 >So, both routers 1 and 2 got the prefix 11.0.0.0/8 >After 1 received 3 data packets, its MED goes to 1. >The quetions is... >If router 3 is going to send packet to 11.0.0.1. It destinate the packet at >router 2. But the packet will still go to router 1 as next hop. When router >1 got this packet, will it forwards it to router 2? >>From s7-1.out, we have "2 11.0.0.0/8 2 1 (med=0)". However, router 1 also >have the prefix 11.0.0.0/8. It should consume the packet logically. >So... should router 1 still forward it? or consume it? > >Thanks. > >regards, >Raymond You need to make a decision based on the routing rules we talked about in class. If you think the decision you make is not clear cut, document your reasoning in your README file. -John Heidemann From johnh at ISI.EDU Mon May 10 17:01:31 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 17:02:33 2004 Subject: [Csci551-talk] question about s7-1.in In-Reply-To: <200405100714.i4A7Ewk00149@vapor.isi.edu> Message-ID: <200405110001.i4B01VJP029299@dash.isi.edu> On Mon, 10 May 2004 00:16:54 PDT, "Jayant V. Rao" wrote: >I have a related question - >Topo: 1-2, 1-3, 2-4, 3-4 >1, 2 manage 11.0.0.0/8. > >Suppose now 1 has med=1, and 2 has med=0. >3 sends a message to 11.0.0.1. it should go to 2, and avoid 1 right? The >correct path is: 3-4-2 >Or is it ok to send 3-1-2? Because this is what seems to be happening below, >sending to 2 thru 1, even though we should avoid 1. > >Clarifications please, >Jayant > > > >-----Original Message----- >From: csci551-talk-bounces@mailman.isi.edu >[mailto:csci551-talk-bounces@mailman.isi.edu] On Behalf Of Raymond Pun >Sent: Sunday, May 09, 2004 11:59 PM >To: csci551-talk@ISI.EDU >Subject: [Csci551-talk] question about s7-1.in > >Hi, > >I have a question. >from s7-1.in >we have >router --> prefix >1 --> 11.0.0.0/8 >2 --> 11.0.0.0/8 >3 --> 13.0.0.0/8 >4 --> 14.0.0.0/8 >5 --> 15.0.0.0/8 >So, both routers 1 and 2 got the prefix 11.0.0.0/8 >After 1 received 3 data packets, its MED goes to 1. >The quetions is... >If router 3 is going to send packet to 11.0.0.1. It destinate the packet at >router 2. But the packet will still go to router 1 as next hop. When router >1 got this packet, will it forwards it to router 2? >>From s7-1.out, we have "2 11.0.0.0/8 2 1 (med=0)". However, router 1 also >have the prefix 11.0.0.0/8. It should consume the packet logically. >So... should router 1 still forward it? or consume it? > >Thanks. > >regards, >Raymond Same answer: You need to make a decision based on the routing rules we talked about in class. If you think the decision you make is not clear cut, document your reasoning in your README file. -John Heidemann From harshits at usc.edu Mon May 10 17:30:06 2004 From: harshits at usc.edu (harshit shah) Date: Mon May 10 17:31:41 2004 Subject: [Csci551-talk] last lsa message Message-ID: Should the last lsa message be the one after the initial flooding, or after the lsa flooding taking place after the MED update? Thanks, Harshit From johnh at ISI.EDU Mon May 10 18:17:56 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon May 10 18:19:30 2004 Subject: [Csci551-talk] last lsa message In-Reply-To: Message-ID: <200405110117.i4B1HufG029915@dash.isi.edu> On Mon, 10 May 2004 17:30:06 PDT, harshit shah wrote: >Should the last lsa message be the one after the initial flooding, or after the lsa flooding taking place after the MED update? > >Thanks, >Harshit It's the last one. Which of those is last-er :-) -John Heidemann From liyuan at pollux.usc.edu Thu May 13 15:42:31 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Thu May 13 15:43:48 2004 Subject: [Csci551-talk] Project B grade sent Message-ID: Dear Class, Grades for Project B were sent. You can find the testcases we used at: http://www-scf.usc.edu/~yuanl/551/projbtestcase.tar If you have any questions, please contact me (also cc the grader rohittri@usc.edu ) ASAP. And please make sure that you compare your outputs with the sample outputs carefully! I won't be able to answer emails from tomorrow afternoon for about one week, so please send emails to our grader(rohittri@usc.edu) for urgent questions during that time, but also CC me a copy. Thanks Yuan From atu011 at earthlink.net Mon May 17 18:32:20 2004 From: atu011 at earthlink.net (Aaron Tu) Date: Mon May 17 18:32:54 2004 Subject: [Csci551-talk] ns-2 CPP_NAMESPACE error In-Reply-To: <200405141900.i4EJ07i17630@gamma.isi.edu> Message-ID: <000201c43c77$fe372c60$830aa8c0@LATUALP> When I tried to rebuild setdest (ns-2), I got the following error. I tried searching the web for an answer, but only to find that this problem is not uniquely mine. I wonder if anyone is familiar with the problem and know how to resolve it. $ make g++ -c -Dstand_alone -DSTL_NAMESPACE=@STL_NAMESPACE@ -o setdest.o setdest.cc In file included from ../../../config.h:54, from setdest.h:5, from setdest.cc:57: ../../../autoconf.h:85: error: namespace `CPP_NAMESPACE' undeclared make: *** [setdest.o] Error 1 Have a great summer! Aaron From johnh at ISI.EDU Tue May 18 08:04:10 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue May 18 08:04:55 2004 Subject: [Csci551-talk] ns-2 CPP_NAMESPACE error In-Reply-To: <000201c43c77$fe372c60$830aa8c0@LATUALP> Message-ID: <200405181504.i4IF4AJr005228@dash.isi.edu> On Mon, 17 May 2004 18:32:20 PDT, "Aaron Tu" wrote: >When I tried to rebuild setdest (ns-2), I got the following error. I >tried searching the web for an answer, but only to find that this >problem is not uniquely mine. I wonder if anyone is familiar with the >problem and know how to resolve it. > >$ make >g++ -c -Dstand_alone -DSTL_NAMESPACE=@STL_NAMESPACE@ -o setdest.o >setdest.cc >In file included from ../../../config.h:54, > from setdest.h:5, > from setdest.cc:57: >../../../autoconf.h:85: error: namespace `CPP_NAMESPACE' undeclared >make: *** [setdest.o] Error 1 > >Have a great summer! > >Aaron Rename STL_NAMESPACE to CPP_NAMESPACE in Makefile.in and reconfigure. -John Heidemann From bshukla at usc.edu Fri May 21 10:50:18 2004 From: bshukla at usc.edu (bhavin shukla) Date: Fri May 21 10:51:55 2004 Subject: [Csci551-talk] Honeybees and the internet Message-ID: <553b37a84c7e.40addeea@usc.edu> Hi guys, I found an interesting article in The Economist and wanted to share it with you all: HONEYBEES want to make honey, while internet host providers want to make money. And the rhyme is not the only thing to link honeybees and internet host providers, who operate vast warehouses full of server computers on behalf of their customers. Sunil Nakrani of Oxford University and Craig Tovey of the Georgia Institute of Technology, in Atlanta, have recently developed what they believe is an efficient way of organising internet servers, by mimicking the behaviour of honeybee colonies. The unpredictability of internet traffic (sudden surges of shoppers or stockmarket trading; the ups and downs of an internet auction) gives firms which run servers quite a headache. To maximise their profit, they have to juggle their computers between different applications in order to adapt to changing levels of demand. Only one web-application can be loaded on to a computer at one time, and switching between applications incurs a penalty of five to seven minutes of downtime while the computer is reconfigured. Honeybees have a similar problem. Patches of flowers vary in quality, so a colony needs to “decide” how many bees will forage at each patch and how long they will forage, in order to maximise the rate of nectar collection. Millions of years of evolution have provided the bees with a pretty good solution. Around one-fifth of the bees in a hive are employed as nectar collectors. Their job is to zip back and forth to flower patches, gathering batches of nectar. On returning to the hive, they transfer their takings to one of the stay-at-home food-storer bees, who then stash it in a honeycomb. A nectar-collecting bee judges how good its flower patch is, relative to the patches being visited by its hive-mates, by seeing how long it takes to find an unemployed food-storer bee. If it takes ages, then the forager concludes that its patch is nothing special, and that most of the other forager bees must have had successful runs, too. But if there is a plethora of food-storer bees ready to take the nectar, then the forager realises that it has struck lucky. Using this information, the forager decides whether its flower patch is worthwhile. If so, it signals to others to follow it back by doing the famous waggle dance. The length of this dance indicates how profitable a flower patch is likely to be. Dr Nakrani and Dr Tovey have exploited the honeybees' strategy and applied it to the problems of internet hosts. By comparing individual servers to foraging bees, and customer requests to flower patches, the two researchers have developed a “honeybee” algorithm for internet-server “hives”. Instead of waggle-dancing, a server produces an “advert”, which it sends to the other servers in the hive. The duration of this advert reflects the importance and profitability of that server's customers. Other servers that read the advert act like worker bees following a waggle dance—judging on the basis of the advert, and of their own recent experience, whether to switch from the customers they are currently serving to the new ones being served by the server that produced the advert. Honeybees and internet servers share similar problems right down to the finest level of detail. For example, the downtime penalty incurred when a computer is switched from one web-application to another can be compared to the trade-off for a honeybee that is switching flower patches—an activity that usually requires several attempts before a profitable new patch is located. To see whether their analogy is useful, Dr Nakrani and Dr Tovey tested the honeybee algorithm against the so-called greedy algorithm currently used as the basis for the allocation of servers by most internet host providers. A greedy algorithm is backward looking. It divides time into fixed periods and allocates servers to customers for a period by working out what would have been the most profitable arrangement in the preceding one. When internet traffic was highly variable, the honeybee algorithm outperformed greedy by as much as 20%. If the traffic became more uniform, then the greedy algorithm began to outperform honeybee. However, most internet traffic is, indeed, highly variable. That suggests Dr Nakrani and Dr Tovey might be on to something. As is often the case with human inventions, nature got there first. Bhavin Shukla Graduate Student in the Computer Science Department University of Southern California http://www-scf.usc.edu/~bshukla/ From naa at usc.edu Fri May 21 11:42:04 2004 From: naa at usc.edu (nishant agarwal) Date: Fri May 21 11:42:24 2004 Subject: [Csci551-talk] Honeybees and the internet Message-ID: <817039727d89.40adeb0c@usc.edu> Please let me know procedure for unsuscribing from the 551 mailing list. I dont like spam on my account. ----- Original Message ----- From: bhavin shukla Date: Friday, May 21, 2004 10:50 am Subject: [Csci551-talk] Honeybees and the internet > Hi guys, > I found an interesting article in The Economist and wanted to share it > with you all: > > HONEYBEES want to make honey, while internet host providers want to make > money. And the rhyme is not the only thing to link honeybees and internet > host providers, who operate vast warehouses full of server computers on > behalf of their customers. Sunil Nakrani of Oxford University and Craig > Tovey of the Georgia Institute of Technology, in Atlanta, have recently > developed what they believe is an efficient way of organising internet > servers, by mimicking the behaviour of honeybee colonies. > > The unpredictability of internet traffic (sudden surges of shoppers or > stockmarket trading; the ups and downs of an internet auction) gives firms > which run servers quite a headache. To maximise their profit, they have to > juggle their computers between different applications in order to adapt to > changing levels of demand. Only one web-application can be loaded on to a > computer at one time, and switching between applications incurs a penalty > of five to seven minutes of downtime while the computer is reconfigured. > > Honeybees have a similar problem. Patches of flowers vary in quality, so a > colony needs to “decide” how many bees will forage at each patch and how > long they will forage, in order to maximise the rate of nectar collection. > > Millions of years of evolution have provided the bees with a pretty good > solution. Around one-fifth of the bees in a hive are employed as nectar > collectors. Their job is to zip back and forth to flower patches, > gathering batches of nectar. On returning to the hive, they transfer their > takings to one of the stay-at-home food-storer bees, who then stash it in > a honeycomb. > > A nectar-collecting bee judges how good its flower patch is, relative to > the patches being visited by its hive-mates, by seeing how long it takes > to find an unemployed food-storer bee. If it takes ages, then the forager > concludes that its patch is nothing special, and that most of the other > forager bees must have had successful runs, too. But if there is a > plethora of food-storer bees ready to take the nectar, then the forager > realises that it has struck lucky. > > Using this information, the forager decides whether its flower patch is > worthwhile. If so, it signals to others to follow it back by doing the > famous waggle dance. The length of this dance indicates how profitable a > flower patch is likely to be. > > Dr Nakrani and Dr Tovey have exploited the honeybees' strategy and applied > it to the problems of internet hosts. By comparing individual servers to > foraging bees, and customer requests to flower patches, the two > researchers have developed a “honeybee” algorithm for internet-server > “hives”. Instead of waggle-dancing, a server produces an “advert”, which > it sends to the other servers in the hive. The duration of this advert > reflects the importance and profitability of that server's customers. > Other servers that read the advert act like worker bees following a waggle > dance—judging on the basis of the advert, and of their own recent > experience, whether to switch from the customers they are currently > serving to the new ones being served by the server that produced the advert. > > Honeybees and internet servers share similar problems right down to the > finest level of detail. For example, the downtime penalty incurred when a > computer is switched from one web-application to another can be compared > to the trade-off for a honeybee that is switching flower patches—an > activity that usually requires several attempts before a profitable new > patch is located. > > To see whether their analogy is useful, Dr Nakrani and Dr Tovey tested the > honeybee algorithm against the so-called greedy algorithm currently used > as the basis for the allocation of servers by most internet host > providers. A greedy algorithm is backward looking. It divides time into > fixed periods and allocates servers to customers for a period by working > out what would have been the most profitable arrangement in the preceding one. > > When internet traffic was highly variable, the honeybee algorithm > outperformed greedy by as much as 20%. If the traffic became more uniform, > then the greedy algorithm began to outperform honeybee. However, most > internet traffic is, indeed, highly variable. That suggests Dr Nakrani and > Dr Tovey might be on to something. As is often the case with human > inventions, nature got there first. > > > Bhavin Shukla > Graduate Student in the Computer Science Department > University of Southern California > http://www-scf.usc.edu/~bshukla/ > > From johnh at ISI.EDU Fri May 21 13:36:48 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri May 21 13:37:18 2004 Subject: [Csci551-talk] Honeybees and the internet In-Reply-To: <817039727d89.40adeb0c@usc.edu> Message-ID: <200405212036.i4LKamr3017966@dash.isi.edu> On Fri, 21 May 2004 11:42:04 PDT, nishant agarwal wrote: >Please let me know procedure for unsuscribing from the 551 mailing list. I dont like spam on my account. I believe that Bhavin's post was because he found it interesting. However, as a reminder for everyone on the list (you should have also gotten this information when you subscribed, and it's in the headers of every post to the list): to unsubscribe, go do http://mailman.isi.edu/mailman/listinfo/csci551-talk or send mail with mailto:csci551-talk-request@mailman.isi.edu?subject=unsubscribe If you need your list password, you can also request it from the above web page. -John Heidemann > >----- Original Message ----- >From: bhavin shukla >Date: Friday, May 21, 2004 10:50 am >Subject: [Csci551-talk] Honeybees and the internet > >> Hi guys, >> I found an interesting article in The Economist and wanted to share it >> with you all: >> >> HONEYBEES want to make honey, while internet host providers want to make >> money. And the rhyme is not the only thing to link honeybees and internet >> host providers, who operate vast warehouses full of server computers on >> behalf of their customers. Sunil Nakrani of Oxford University and Craig >> Tovey of the Georgia Institute of Technology, in Atlanta, have recently >> developed what they believe is an efficient way of organising internet >> servers, by mimicking the behaviour of honeybee colonies. >> >> The unpredictability of internet traffic (sudden surges of shoppers or >> stockmarket trading; the ups and downs of an internet auction) gives firms >> which run servers quite a headache. To maximise their profit, they have to >> juggle their computers between different applications in order to adapt to >> changing levels of demand. Only one web-application can be loaded on to a >> computer at one time, and switching between applications incurs a penalty >> of five to seven minutes of downtime while the computer is reconfigured. >> >> Honeybees have a similar problem. Patches of flowers vary in quality, so a >> colony needs to ?decide? how many bees will forage at each patch and how >> long they will forage, in order to maximise the rate of nectar collection. >> >> Millions of years of evolution have provided the bees with a pretty good >> solution. Around one-fifth of the bees in a hive are employed as nectar >> collectors. Their job is to zip back and forth to flower patches, >> gathering batches of nectar. On returning to the hive, they transfer their >> takings to one of the stay-at-home food-storer bees, who then stash it in >> a honeycomb. >> >> A nectar-collecting bee judges how good its flower patch is, relative to >> the patches being visited by its hive-mates, by seeing how long it takes >> to find an unemployed food-storer bee. If it takes ages, then the forager >> concludes that its patch is nothing special, and that most of the other >> forager bees must have had successful runs, too. But if there is a >> plethora of food-storer bees ready to take the nectar, then the forager >> realises that it has struck lucky. >> >> Using this information, the forager decides whether its flower patch is >> worthwhile. If so, it signals to others to follow it back by doing the >> famous waggle dance. The length of this dance indicates how profitable a >> flower patch is likely to be. >> >> Dr Nakrani and Dr Tovey have exploited the honeybees' strategy and applied >> it to the problems of internet hosts. By comparing individual servers to >> foraging bees, and customer requests to flower patches, the two >> researchers have developed a ?honeybee? algorithm for internet-server >> ?hives?. Instead of waggle-dancing, a server produces an ?advert?, which >> it sends to the other servers in the hive. The duration of this advert >> reflects the importance and profitability of that server's customers. >> Other servers that read the advert act like worker bees following a waggle >> dance?judging on the basis of the advert, and of their own recent >> experience, whether to switch from the customers they are currently >> serving to the new ones being served by the server that produced the advert. >> >> Honeybees and internet servers share similar problems right down to the >> finest level of detail. For example, the downtime penalty incurred when a >> computer is switched from one web-application to another can be compared >> to the trade-off for a honeybee that is switching flower patches?an >> activity that usually requires several attempts before a profitable new >> patch is located. >> >> To see whether their analogy is useful, Dr Nakrani and Dr Tovey tested the >> honeybee algorithm against the so-called greedy algorithm currently used >> as the basis for the allocation of servers by most internet host >> providers. A greedy algorithm is backward looking. It divides time into >> fixed periods and allocates servers to customers for a period by working >> out what would have been the most profitable arrangement in the preceding one. >> >> When internet traffic was highly variable, the honeybee algorithm >> outperformed greedy by as much as 20%. If the traffic became more uniform, >> then the greedy algorithm began to outperform honeybee. However, most >> internet traffic is, indeed, highly variable. That suggests Dr Nakrani and >> Dr Tovey might be on to something. As is often the case with human >> inventions, nature got there first. >> >> >> Bhavin Shukla >> Graduate Student in the Computer Science Department >> University of Southern California >> http://www-scf.usc.edu/~bshukla/ >> >>