From chrisramming@yahoo.com Thu Oct 2 19:20:19 2003 From: chrisramming@yahoo.com (J. Christopher Ramming) Date: Thu, 2 Oct 2003 11:20:19 -0700 (PDT) Subject: [KP-seed] briefing review for Tues. 10/7/03, 2pm EDT Message-ID: <20031002182019.80661.qmail@web41511.mail.yahoo.com> Hi Folks, I had hoped to convey some concrete news at our next status meeting, but I don't have any yet after all. So rather than have one of our usual status meeting on 10/7 @ 2pm EDT, I plan to distribute a planned briefing on that date and walk through it for anyone who is interested in reviewing it. This will be an informal and optional working call on the usual bridge number: 800.771.7760 participant code 620599. Best regards, Chris From chrisramming@yahoo.com Tue Oct 7 04:47:00 2003 From: chrisramming@yahoo.com (J. Christopher Ramming) Date: Mon, 6 Oct 2003 20:47:00 -0700 (PDT) Subject: [KP-seed] briefing materials for discussion tomorrow Message-ID: <20031007034700.70259.qmail@web41501.mail.yahoo.com> folks, rather than risk another list server problem, i've placed the briefing materials for tomorrow's discussion in: http://f1.pg.briefcase.yahoo.com/chrisramming this obscenely large powerpoint presentation (even though compressed) contains a combination of 1. background on tactical network management 2. briefing slides addressing heilmeier questions 3. slides excerpting relevant points from each of your seedling efforts as they apply to proposed work in a cognitive networks program because i think that the key risk of a program like this is the architecture (which is tightly bound to the kinds of learning and reasoning about networks that get invented), i've presented a bunch of your results as architectural explorations even though i could have put together a slide about how your work informs one of the more detailed technical topics too. if you would prefer to present your results with a different slant please feel free to provide an alternative version for me to include. finally, regarding those slides about your individual projects: in summarizing such substantive works in a few short bullet lists i'm doubtless doing a disservice to your insights and results. please help me correct oversights and so on either by providing a revised version of the slide (without departing from the structure if possible) or by sending me an email containing requested corrections. best regards, chris From chrisramming@yahoo.com Fri Oct 10 01:09:34 2003 From: chrisramming@yahoo.com (J. Christopher Ramming) Date: Thu, 9 Oct 2003 17:09:34 -0700 (PDT) Subject: [KP-seed] final (?) draft Message-ID: <20031010000934.52425.qmail@web41505.mail.yahoo.com> http://briefcase.yahoo.com/chrisramming contains a new version that i think/hope reflects all of your comments as well as i have been able to incorporate them. the biggest changes in the last few days concern new input about tactical network issues and are reflected largely in the first section. there's also a new diagram to express an idea about how the cognitive plane relates to the management plane (thanks to bob marcus of SRI for finding a way to depict a solution). in the event anyone sees glaring problems please email me at my blackberry account "jramming@snap.org", otherwise have a good weekend (i'm on vacation until monday and won't be checking this account if willpower triumphs). thanks to all of you for your hard work. this is an important thing for DARPA to do, and each of you has done a lot to make the case for a program in this area so i hope i'll be able to make that case effectively. From fmartin@cs.orst.edu Wed Oct 15 06:48:02 2003 From: fmartin@cs.orst.edu (Francisco J Martin) Date: Tue, 14 Oct 2003 22:48:02 -0700 (PDT) Subject: [KP-seed] IBM, Cisco look to help networks help themselves Message-ID: i've just received this piece of news: IBM, Cisco Look To Help Networks Help Themselves The two companies are teaming up to find ways of helping companies solve problems in their IT infrastructures. http://www.computerworld.com/newsletter/0,4902,85947,00.html?nlid=NET best wishes, Francisco From chrisramming@yahoo.com Sat Oct 18 18:37:19 2003 From: chrisramming@yahoo.com (J. Christopher Ramming) Date: Sat, 18 Oct 2003 10:37:19 -0700 (PDT) Subject: [KP-seed] a ray of light Message-ID: <20031018173719.10506.qmail@web41503.mail.yahoo.com> Hi All, Just a quick note to update you on yesterday's briefing. There were seven people present: Tony Tether, Bob Leheny, Dave Honey, Barbara Yoon, a Col Johnson (associated with FCS), Jeanette Milos, and Ron Brachman. Give the concurrent DARPA picnic that was not a bad turnout. In slightly over one hour of discussion I managed to get through a full abbreviated presentation (uploaded to briefcase.yahoo.com/chrisramming) of 12 slides about the program idea, and even some detailed discussions of your individual projects. I managed to stay in the higher ends of my personal presentation range and feel confident that I accurately represented what we have done and what should be done next. In particular, there was no ambiguity about the technology vs. systems program distinction and the idea that we are proposing a foundational technology program in cognitive networking. Therefore, anything other than a negative outcome is probably a good sign (but not a definitive one). Tony asked me to meet with him again, and I will be going back to Washington to have an early morning followup on Tuesday. Anything could happen, but I think it is clear that the importance and value of the work you have done so far has been recognized. Thank you all for the hard and excellent work you've done, and for your help in getting the project to this stage. Chris From jramming@snap.org Mon Oct 20 22:26:57 2003 From: jramming@snap.org (jramming(contr-ipto)) Date: Mon, 20 Oct 2003 17:26:57 -0400 Subject: [KP-seed] No status meeting tomorrow Message-ID: <32D3609E0ED18D4BA8EA845B3C26CA4002DBED53@exchange5.darpa.mil> By your leave I would prefer not to have a status meeting until I have something definitive to report that is actionable by the group. Let's hope and expect the next meeting to be on 10/28. Thanks Chris -------------------------- Sent from my BlackBerry Wireless Handheld From tgd@cs.orst.edu Fri Oct 24 23:02:28 2003 From: tgd@cs.orst.edu (Thomas G. Dietterich) Date: Fri, 24 Oct 2003 15:02:28 -0700 Subject: [KP-seed] [KP] International Conference on Autonomic Computing (ICAC-04) Message-ID: <3716-Fri24Oct2003150228-0700-tgd@cs.orst.edu> IBM et al are organizing a conference in NY next may in conjunction with the WWW conference. You can read about the details at http://www.autonomic-conference.org Paper deadline is January 12, 2004 --Tom -- Thomas G. Dietterich, Professor Voice: 541-737-5559 School of Electrical Engineering FAX: 541-737-3014 and Computer Science URL: http://www.cs.orst.edu/~tgd Dearborn Hall 102, Oregon State University, Corvallis, OR 97331-3102 _______________________________________________ Know-plane mailing list Know-plane@mailman.isi.edu http://mailman.isi.edu/mailman/listinfo/know-plane From braden@ISI.EDU Mon Oct 27 04:42:03 2003 From: braden@ISI.EDU (Bob Braden) Date: Sun, 26 Oct 2003 20:42:03 -0800 (PST) Subject: [KP-seed] Some of the competition... Message-ID: <200310270442.h9R4g3K08152@boreas.isi.edu> "Smart Servers as Watchdogs for Trouble on the Web" New York Times (10/23/03) P. E8; Eisenberg, Anne An upgraded Internet that can detect worms, traffic bottlenecks, and other network problems before they become serious may one day be within reach thanks to the efforts of PlanetLab, an academic-industrial consortium that has created a virtual testbed network built atop the Internet. PlanetLab employs PCs as smart routers at each network node, and these machines can run applications designed to detect whether data packets are benign or malign. "We are putting all the functionality and smarts into the PC at each node, without disturbing the rigid capabilities of the Internet routers," explains PlanetLab leader and Princeton computer science professor Larry Peterson. Hewlett-Packard, Intel, and Google have contributed hardware and services to the project. By donating nodes at their sites, researchers can use the PlanetLab network to test their tools as well as leverage the broader scope of network services; one tool, a worm detector called Netbait, is being tested on PlanetLab by Intel scientist Brent Chun, who says the tests have yielded new insights on worms and the spread of their infection. PlanetLab is also being used to test CoDeeN, a program developed by Princeton researchers to mitigate network traffic jams by replicating popular Web pages and positioning them on servers near potential requesters. PlanetLab is purchasing more machines to bolster the network infrastructure using funds from the National Science Foundation, and Peterson reports that the network is currently linked to 2,600 of the Internet's approximately 15,000 autonomous systems. About 250 PC nodes have been deployed so far, and Peterson says 1,000 nodes--enough to provide full coverage--should be set up in two years. From chrisramming@yahoo.com Mon Oct 27 15:33:00 2003 From: chrisramming@yahoo.com (J. Christopher Ramming) Date: Mon, 27 Oct 2003 07:33:00 -0800 (PST) Subject: [KP-seed] status call 10/28 2pm edt Message-ID: <20031027153300.75898.qmail@web41502.mail.yahoo.com> Please let me know if any of you have done work you'd like to report on in this call. -------------------------------------------------------- Tuesday, 6/10/2003, 2pm EDT Conference bridge: 800-771-7760, participant code 620599 1. review of agenda 2. jcr to give status update on program status From jrex@research.att.com Mon Oct 27 15:50:44 2003 From: jrex@research.att.com (Jennifer Rexford) Date: Mon, 27 Oct 2003 10:50:44 -0500 (EST) Subject: [KP-seed] position statements and slides from routing workshop Message-ID: <200310271550.KAA40313@chips.research.att.com> KP folks, A few weeks ago I co-ran a small workshop on open issues in the evolution and design of Internet routing protocols http://www.net.informatik.tu-muenchen.de/wired The participants (researchers, operators, and vendors) submitted two-page position statements and five-minute talks. We have the statements and slides online at the Web site. Several of the statements are germane to issues we've been discussing in the KP seedling about network configuration and troubleshooting. -- Jen From peyman@MIT.EDU Thu Oct 30 20:50:44 2003 From: peyman@MIT.EDU (Peyman Faratin) Date: Thu, 30 Oct 2003 15:50:44 -0500 Subject: [KP-seed] User Level Internet Path Diagnosis Message-ID: --============_-1144575451==_ma============ Content-Type: text/plain; charset="us-ascii" ; format="flowed" Hi This paper presented at this year's Symposium on Operating Systems Principles (SOSP-03) may be of interest to some people. http://www.cs.rochester.edu/sosp2003/papers/p228-mahajan.pdf Abstract: Diagnosing faults in the Internet is arduous and time-consuming, in part because the network is composed of diverse components spread across many administrative domains. We consider an extreme form of this problem: can end users, with no special privileges, identify and pinpoint faults inside the network that degrade the performance of their applications? To answer this question, we present both an architecture for user-level Internet path diagnosis and a practical tool to diagnose paths in the current Internet. Our architecture requires only a small amount of network support, yet it is nearly as complete as analyzing a packet trace collected at all routers along the path. Our tool, tulip, diagnoses reordering, loss and significant queuing events by leveraging well deployed but little exploited router features that approximate our architecture. Tulip can locate points of reordering and loss to within three hops and queuing to within four hops on most paths that we measured. This granularity is comparable to that of a hypothetical network tomography tool that uses 65 diverse hosts to localize faults on a given path. We conclude by proposing several simple changes to the Internet to further improve its diagnostic capabilities. Peyman -- ______________________________________________________________________________ Peyman Faratin: MIT Laboratory for Computer Science and Artificial Intelligence 200 Technology Square, NE43-534, Cambridge, MA, 02139 Tel: +1 (617) 258-0458, Fax: +1 (617) 253-2673 http://ana.lcs.mit.edu/peyman/ email: peyman@mit.edu ______________________________________________________________________________ --============_-1144575451==_ma============ Content-Type: text/html; charset="us-ascii" User Level Internet Path Diagnosis
Hi

This paper presented at this year's Symposium on Operating Systems Principles (SOSP-03) may be of interest to some people.

http://www.cs.rochester.edu/sosp2003/papers/p228-mahajan.pdf


Abstract:

Diagnosing faults in the Internet is arduous and time-consuming,
in part because the network is composed of diverse components
spread across many administrative domains. We consider an extreme
form of this problem: can end users, with no special privileges,
identify and pinpoint faults inside the network that degrade
the performance of their applications? To answer this question,
we present both an architecture for user-level Internet path diagnosis
and a practical tool to diagnose paths in the current Internet.
Our architecture requires only a small amount of network support,
yet it is nearly as complete as analyzing a packet trace collected
at all routers along the path. Our tool, tulip, diagnoses reordering,
loss and significant queuing events by leveraging well deployed but
little exploited router features that approximate our architecture.
Tulip can locate points of reordering and loss to within three hops
and queuing to within four hops on most paths that we measured.
This granularity is comparable to that of a hypothetical network
tomography tool that uses 65 diverse hosts to localize faults on a
given path. We conclude by proposing several simple changes to
the Internet to further improve its diagnostic capabilities.


Peyman

--
______________________________________________________________________________

 Peyman Faratin: MIT
 Laboratory for Computer Science and Artificial Intelligence
 200 Technology Square,  NE43-534, Cambridge, MA, 02139
 Tel: +1 (617) 258-0458,  Fax: +1 (617) 253-2673
 http://ana.lcs.mit.edu/peyman/ email: peyman@mit.edu
______________________________________________________________________________
--============_-1144575451==_ma============--