^ Top

NANOG 23 Online Evaluation Results

Oakland, CA 

Number of responses:  150 


(# Res. = Number of responses) (Avg. = Average from responses) (Scale: 1 = Excellent, 5 = Unsatisfactory) 









# Res.


# Res.


# Res.



1:30 PM



IP Routing Protocol Scalability









ISP Security








3:30 PM



IP Routing Protocol Scalability









ISP Security








7:30 PM



BGP Multihoming Guide









Packets and Photons









Breakout Sessions

(# Res. = Number of responses) (Avg. = Average from responses) (Scale: 1 = Excellent, 5 = Unsatisfactory) 







 7:30 PM


# Res.


# Res.


# Res.




Data Center Peering









Testbed for Evaluating DDoS









Other Questions 





1st time at NANOG?








Your organization interested in hosting?



Contact Info

Why are you here? 




# Responses


For Operators





See new hardware?




Discuss new knobs?




Hear BCP presentations?




Hear backbone technology presentations?




Meet other operators?




Meet other vendors?











For Vendors





Meet customers?




Gauge demand for product?




Quietly pitch product?




Hear backbone technology presentations?











Do you have suggestions for future NANOG presentations?  (Topics, and/or speakers) 

What worked well and what should be improved for the next NANOG? 

What worked well and what should be improved for the next NANOG? 

Great facilities and terminal root. The wireless cards were a nice touch aswell.A single laptop should be used for all presentations. It would be nice to have the beer and gear earlier so that you can meet more people. 


wireless was very good.Did not even half way feel the need to go to the terminal room.(It was nice knowing it was there as a backup, tho) 


The wireless was having trouble, too much traffic on the access points. Would really appriciate that the room has more AP and that some form of load balancing between the APs could/may/might help. 


Too many DDOS presentations. 
It would be good to have the Monday lunch catered by a sponsor.The first day in a new city makes it awkward to know where to go and half of the lunch break time is wasted chasing food locations and fighting with the elevators to stow computers or grab jackets. With the old system you met new people as the tables got filled in.This system keeps the folks who know each other in groups and alienates the new people.If it can't be catered by a sponsor in full/part, collect 10-15 bucks at the door.Removing the hassel of lugging computers and stuff around random streets would be worth it. 


Every thing works very well. 


Really like the wireless WEB access and most of the presentations being on line during the presentations.Liked the WEB cast a lot as well.But, it would be nice to have recordings available of the sessions. 


Wireless access was terrible. Even being next to an access point was not enough to have reliable and non-latent connectivity. 


more wireless access points 


wireless needs to be improved - packet loss from wireless aggregation point to fist hop upstream was 25% from my measurements 


Wireless network was slooowwwww....  Nice facilities, lots of room! 


Cisco's (the vendor who makes the equipment) wireless coverage was really poor. 


The venue itself was nice, but who picked Oakland?

The wireless config was sub-optimal, but at least there were some wired ports available.

I would like to possible see us cut the sunday night tutorial and maybe have more meaty topics. 


As usual, the meeting was well put together. The agenda worked well, the hotel worked well, the wireless network worked well most of the time. The meeting room was very good. Keep up the good work. 


Location was great didn't like Cisco putting Oakland down in the intr.Could be because I was born here. Network and availability was good. Conf rooms good, bar needed more chairs!

Tutorials on the core protocols are always welcome. Like keeping the market droids away from the conference as best as possible.

Updates on standards is great 
(bill W talk was great)

Updates from research community 
also great. CAIDA, Merit,USC, etc.

Beer and Gear went well. 


Room was way too cold.  Rest was very good 


Wireless connectivity has been very bad.  Design problem? 


Multiple screens and many tables are good. 
Suggest Hawaii as a future venue. 


"TeraPOP" panel was a joke.Very little content about terabit or even large POPs. 
Only 1 presenter, Sprint, even came close. 


too cold! 
bad connectivity (losing lots of packets) 
good quick registration 
good local info (restaurants,...) 
nice, interesting talks 


Everything was first rate.Last time I was here was '97 in Scottsdale.It has grown! 


the wireless network seemed saturated 
too often? 


Everything went fairly well this time, espically given the loss of Abha and the need for shuffling. 


Wireless connectivity in the main conference room could've been improved at times when throughput was abysmal. 


The conference itself was laid out and organized well. The agenda worked well despite the last minute changes.

I didn't like the deal w/ the women's bathrooms, but I understand that the hotel decision and not an organizational decision from the hosts.

It is a sad time for all of us, plus the world is a different place. I wonder how that will affect future conferences. 


On this NANOG, I liked the tutorials, which were at a good intermediate level and were a good learning experience.

On the other hand, I found the general session not so useful.There were too many academic talks on DDoS, and too little practical presentations on useful operator's issues. 


ALL talks slides should be on web before theyre presented. 
Kireeti, Ogielski 's talks for instance didnt.

archive av recordings for a few months. 


Things are improving. Most things worked well. 


What worked? 
The wireless network

What should be improved? 
All presentations were not available online. 
Agenda haphazardly rescheduled. People interested in attending specific presentations were disappointed. 


Connectivity.I was providing a repeated on the 7th floor which was providing connectivity to many rooms.Monday night, the repeater on the second floor I was associated with went away.This caused loss of connectivity for many people.

These are net-geeks here.They want to have fun and have net access, so :) 


table layout and space for walking between them.wireless and presentation screens worked very well also. 


Wireless access was great. Facilities were good. Break times were appropriate - sufficient time to mingle.

Overall, a very nicely organized conference! 


consistant quality of video and audio. 
I know you can't totally control slides and whatever the speakers come with, but having audio black holes in the room where video screen block speakers and the roar of the hallway chatter drowns the speakers especially when they speak softly and noone is riding their mile gain to compensate is just sloppy management or poor vendor selection. You pay them, demand quality.

Speaker and mike placement, compensation for room acoustics, active control for changing conditions all make the difference between easily being able to hear what is being said without being aware of the "sound system" and being TOTALLY aware of the inadequate "sound system"

Having rude blaring distorted "MUSIC" intrude on break time conversations with fellow attendees again makes the "SOUND SYSTEM" noticable.

A good sound system isn't noticed - it just does its job very well. 


Presentations went pretty smoothly and didn't drone on. 


WLAN support is not as good as it could be. Esp. on Monday I couldn't use it at all. DHCP would be great (though you then would also definetly have to have a mailrelay ... or catch SMTP traffic) 


Wireless Lan rocks 


one improvement..Make the company names larger on the badges so that they can be read from futher away.. 


glad to see us moving in the direction of more wireless and less wired network connectivity.

the question address didn't seem to be well defined and lots of people were emailing complaints about temperature, webcast problems, etc. also, it seems it would best if someone who was sitting in the front of the room were to handle asking questions from the list, since it is almost impossible to interrupt the discussion to use the mic to ask questions from people who emailed the question list. 


Don't start so early in the morning; even the east coasters are bleary. 


It would be great if we could get *all* of the presentations in softcopy. The pace of presentation is too fast to take extensive notes. 


having complete wireless access along power outlets and video screens throughout the meeting area have always been extremely helpful. 


I wish all the presentation would be available online. 


Get the slides on-line before the talks!

Also, the remote deliverly seemed to work quite well for those of us who can't travel.



What worked: Overall well put together.

What Didn't:Numerous Cisco Marketing plugs.Also some of the DDOS lectures were a little redundant, but I did enjoy the DDOS/Worm solutions.

Improvements:Perhaps having a few people speak on problems they've seen along their solution to the problem/issue with real life examples :) 


improve wireless connectivity. 
need denser 802.11b access point coverage. 
maybe give 802.11a a try as well. 


I enjoyed it. 


802.11b wireless rocked! 


I really appreciate that all the presentations are in a single room and we all get to see them. 
The sharing of knowledge and ideas is excellent. 
Wireless access rocks. 
Multicast is great! 
Looks like lighting should be worked out for the presenters. 


Nice hotel, fair location. Very good conference rooms. Need to get multi-channel 802.11b and lock down lower speeds on the network for more reliable connectivity. 


(P.S. to my earlier eval)

Many of the presentations showed charts with unreadable lines and/or axis labels; these were especially frustrating because many of them were unavailable for download in softcopy. Please provide all presentations in softcopy and ask presenters to make charts more readable. 


Many IPv6 routes were disable at NANOG. 
ex. 3ffe:8086/?? 
Please care about IPv6 routing.

Thanks for your great works!! 


Don't support 2Mb/s wireless cards - get more comfortable chairs/seats 


I enjoy the flexibility and semi-informality of the program.

I liked the three presentations on 9/11 as well as the Tuesday BGP papers. 


The room was freezing.

Closing all of the women's restrooms on the ground floor because there are more men than women at the conference was a stupid idea -- I'm glad it was aborted early. 
(and incidentally, I never saw a line at a men's room during the conference)

Next time maybe have more ice cream at the break -- I didn't get any :( 


The Q&A format was not good, this time. Give and take is really essential for the Q&A 


Data Center Peering BOF was good. 
Microphones should be on in the audience. 
Keep the meeting interactive don't hold question til the end or through email. 


NANOG:In my opinion, everything worked well.There was nothing that I felt needed improvements. 


All was ok. 


Improve wireless network performance.

The short Network Management tool presentations were very valuable. 


The BGP presentation track on Tuesday was a very effective forum for concentrating on the current issues facing the Internet community.We had presentations from several groups working independently, and hopefully the forum will promote better better collaboration, improving everyone's understanding. 


I was somewhat disappointed the terminal room was closed Monday night.:(It was mitigated by the fact the wireless stayed up so I could sit out in the hallway and plug into a wall outlet with my laptop.But please keep in mind, we are geeks, and like to stay connected regardless of the hour.:) 


The audio was very good this time, and it is often problematic. Most of the presentations were really good this time, too. The wireless network seemed to have performance problems often. I am not sure the emailed questions worked all that well as the *only* way to handle questions, but I am in favor of keeping it as an option. We thanked a lot of people; did we thank Susan?Thanks, Susan!(I can't remember) 


it would be better if all presestantations were available at time of presentation if not soon after. 
it also seems like a good idea to have them displayed form one platform ie no laptop changes between speakers. 
also provide a mike that is attached to speaker. 
otherwise an interesting set of presentations. 


This one had some talks too vendor oriented. 


DDOS needs to be ddos'ed. It's been killed already. Stop beating it!

The Sept. 11th (what worked, what didn't) was a good overview. 


Very impressed, have not attended NANOG meetings for several years - last 1998. 


NANOG should switch to an elected (by the attendees, once a year) committee for setting the agenda for each meeting.The current group are giving us the same tired subjects every time, and some of them have significant conflicts of interest.They are not representative of the group as a whole. 


Wireless LAN worked great. The DNS service sucked, though. Get a decent Linux box for DNS the next time... 


Better wireless coverage would be nice, things appeared a bit stretched this time around. 


I'm am *sick* and *tired* of DDOS and MPLS presentations.NANOG has totally overdone them, and they don't really give an answer, especially DDOS.

It really ticked me off when we had DDOS and MPLS for most of Monday, when everyone was at the conference, and then had the more informative presentations on Tuesday (especially MOAS, The Impact of BGP misconfiguartion and the other topics later in the day) when *many* people were not around because they wereleaving on Tuesday.

Frankly, the presentations on Tuesday would/can potentially stop quite a few flame wars on NANOG, which is always needed.:)

Please, if NANOG is going to present more DDOS and MPLS presentations, please have them on Tuesday.

Another thing I have noticed is that quite a few interesting-sounding presentations are done in the early AM (e.g. - Internet under stress, What worked what didn't 9/11, Multicast worked on 9/11).I know quite a few people, including me, have trouble waking up that early in the morning, whether it jet lag, different sleeping schedules, or what have you.

This may sound vague, but if we could have the "more interesting" presentations in the afternoon that seem to be in the morning, I personally would appreciate it. ;-) 


The registration went smoothly and the food was good.

The wireless network had a lot of problems: high latency, mysterious outages, dns outages, blecch.  Very frustrating.

I really liked the music during the breaks!

The t-shirt was pretty boring.  tie dye or color variation would have been a big improvement.  Grey with a bit of text doesn't excite me.

The power drops and tables in the main room were *awesome*.  Great job on that.  The four projectors and screens was great as well. 



Suggestions for future NANOG presentations? 

Netflow based analysis techniques. 


More optical routing. 


Have Mike Hughes (LINX) or Fearghas McKay lead a panel of NAP operators.WCOM MAEs, PAIX, Equinix, SBC, Switch&Data, Telehouse, Sprint, etc.Panelists should have a list of items to report to: 
1. Locations active and planned 
2. Peering product offered 
3. Link to website with list of participants 
4. Link to website with outage details 
5. Upgrades in progress for coming 6 months 
NANOG participants can better use this information from a NAP operator to plan for new exchanges, locate peers and keep contacts up to date in this rapidly changing industry. 


Real Carrier Grade Networks 


Would like to see more on methodoligy and recent work regarding issues of the day and future. 
Even how other people solved some of these issues.Wether or not we like it or not makes no difference. 


Test Tools, simulations 


I'd like to see other domestic U.S. carriers other than sprint discuss their network design and operating practices. 


Configuration management practices and tools

Building exchange points: how did "new" exchange points like Atlanta and NAP of the Americas get off the ground? 


More real world presentations.Allow speakers a longer period of time to prep.Otherwords start making selectons more than 45 days before the meeting.That way speakers can mix the additional work into their already busy schedules. 


if 47% of the attendees are operators, why-o-why do we still spend hours talking about ddos attacks?

The network operators live routing, h/w configs, etc. Learning what the kiddies are doing is fine, but put them at the end of the agenda.

It's to a point that I don't even come for the presentations any more, there is usually only one good talk about networking for both days.

I also don't want to see presentations from some researcher who hasn't logged into a router ever! I.E. non-operators shouldn't be giving us presentations. 


Considering DNSSEC is close to being deployed, a state of the onion from the VeriSign folks would be nice.Talks regarding routing protocol security would be good to see. 


We do need some more good quality operational content - sharing of experiences, new tools, etc.

This has been one of the best meetings for a long time! 


Prefer presentations that present a problem and how it was solved. Old problems ok if the solution is novel or based on what new technology does. 


Effect of emerging Route Optimization products for multi home Ases 


MAN sized 802.11 deployment - Tim Pozar of BAWUG 


more BOFs and although it is supposed to be NANOG, a more international flavor (possibly tracked) would be welcome -- there is far too little interaction between APNIC and AFNOG participants and NANOG ones...hard to coordinate, but would be very valuable. 


Dump some of the DDOS topics. 


The topics were great as always.Some of the speakers didn't use the microphone adequately or enunciate clearly which made it difficult to hear and understand some of the more interesting parts of their talks. 


NAP connection statistics,movement, and new changes. 


applicaitons of network simulation to mock up real nets 


I think it ran very well. 


Something from the content provider persepctive 


Try to not have extreamly simular presentations.It gets rather boring. 


discussion of current legislation that was passed after sept 11th. 


More Optical-Electrical 
More MPLS 
Peering Acceptance Criterion Analysis (ie: how to verify ISP X is handling 1G/sec, how many nets do they host, etc.) 
DNS interworking (mapping start of authority) 
BGP extensions being worked on & attributes & IBGP route shaping techniques 


I think a presentation on current trends in Peering would be useful, specifically with regard to the much-debated move of the top Tier-1 providers into common space and the new options available to Tier-2 and Content providers. I'm working in this area and will submit an abstract next time. 


BOFs every night, evening roundtables in informal form. 
Employment Board, whos looking and whos offering. 


Please please please make sure the presentations are available on line BEFORE the speakers talk.

For those of us not at the NANAG (watching via Multicast and/or Real) and for those who are visually impaired, it is almost impossible to see the slides. 


Metro case studies, More solutions presentations, In depth MPLS. 


MPLS and DDoS are played and do not warrant new discussions. 


More discussions on new technologies to solve current problems. 


NLRI is not be-all, end-all. Some route optimizations presentations would be cool.

Andy Ogielski's presentation very interesting and thought-provoking.

More on nascent technologies would be cool. 


Intermediate tutorial sessions on routing, down to basics of configuration. 


No more DOS! 


profitability of various services like 2547bis, transit, hosting, etc. 


more on protocols other BGP 


More on BGP configuration and best practices 


More talks on software systems operators use to configure and manage network LANs, WANs, Customer VPNs, etc. 




More DDOS, more CIDR growth table info, more MOAS findings. 


Patrick raised the performance impact of long prefix advertisements.I think this would be nice to see: 
1)Presentation on what one research group has found on a Monday 
2)A BOF to allow all the bickering which is sure to follow.

A good extension of this would be a BCP day time presentation of community architectures to support scoped propogation of advertisements.aka inter-as TE routes leak far enough for the intended effect of the advertiser.

Probably a tutorial but maybe a presentation on what non-traditional methods smaller(more mobile aka non-CLEC) providers are doing to bring customer connections on-net.Things like wireless local loops, "POPs" in multiple tenant buildings, ...Good way to attract or keep the smaller regionals. 


Not yet, but if I do, I'll mail you.:) 


this would require some thought. who should I send suggestions to if I think of some? 


More diversity, this one had lots of DDOS talks. 


The general feeling among people I talked to seemed to be that there were far too many DDOS presentations both this time and last time.While it is certainly an important subject, those of us with other interests tend not to have the attention spans for quite that much of it.

I would like to see a bit more "operational content."As stated earlier, Bill's peering presentations are always welcome.If anybody else is doing peering research, that would be interesting as well.I was quite interested in the BGP presentations on Tuesday, which seemed to be studying some interesting questions, although I wish they had worked harder to draw conclusions rather than just presenting so much raw data.The ARIN session was useful this time.The Network Solutions and exchange point updates, which have been missing from the last few NANOGs, used to be quite useful and it would be nice to get them back. 


This meeting was focused on WTC, DDOS and Routing stability.It be nice to hear quick general status of what's happening on the 'net, Internet2, vbns, etc.

Have Susan Harris, Merit, post more statistics on who is attending, companies/networks, countries, job functions, etc.Not just pretty pictures.Don't have to present but post analysis numbers posted.Shouldn't be too difficult. 



We do need status updates from major exchange operators including Equinix, PAIX, Worldcom/MAE, SprintNAP, Pacbell NAP, AADS NAP, LYNX, etc. 


More focus on the host (such as Linux and BSD). More presentations about stuff like the latest goings-on in the IETF. More IPv6 focus. 


More BGP analysis presentations. 


More indepth large-scale IP routing implementation examples. 
Complex routing and "how not to break the Internet" scenarios would be nice.The BGP sessions were excellent, another BGP type talk directed at a more advanced audience would have been great. 
Some more Case Study presentations for complex or unique routing solutions.(I imagine this would need a bit of time for QA & disection.) 
(I really liked the TUNDRA presention.Very useful, I'd really like to see the code.) 



Packets and Photons


The speaker didn't have detailed knowledge of the presentation. 


Lack of experience in optical networking from presenter 


The slides were weak in detail and the primary speaker were was not hereto give the talk so this great topic was botched. Felt more like a Juniper marketing pitch. 


Substitute speaker skipped or couldn't explain significant parts of content. 


The talk was a bit more theoretical than I preferred.If it was a little more introductory I would've been able to follow it better. 


The speaker was not able to address most questions from the audience, which were very relevant. 


no slides available 


Speaker could not answer technical questions. Looked like a marketing pitch rather than a serious presentation. 


Great! Though I didn't understand everything I got a good idea of what's in the pipe. 


While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


I really wish we'd had the two back-to-back sessions for this, there was so much more detail I would have *loved* for them to have been able to delve more into depth on many of the points. Having the slides up would have been very nice to follow along, too, given the shorter time frame. 


the presentation stayed at a pretty high, abstract level.Would be more informative if diving down a little more. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 


The presentation was too dry, and was too much of a marketing pitch. More coverge of other competing technologies such as QoS routing could have made the presentation more worth while. 


I wish i had gone to the packet and photons one, it sounded interesting. 



Why are you here?

For Operator Comments: 

A mix of: hardware, BCP's, technology, ideas, friends. 


Peering.Contacts, NAP information. 


More diverse subject matter.Too much MPLS and Ddos talks 


Just to learn as much as possible, and meet as many of the NANOG Regulars as possible. 


Meeting people, getting business done, having beer, learning new things 


To contribute to the NANOG community through presentation on security trends. Also to meet people and remain in touch with issues facing network operators. 


I came to see what new hardware is supporting, what other operators are wanting, and what is changing in our evolving networks and fundamental designs. 


meet perring people 


Nanog is the best confernece to go to meet other operators 


Rub elbows trade war stories etc. 


This isn't a question you can select one answer to.Myself I came here to try to put some faces to names.Meeting people you work with is a great thing.

In addition, I came here to attempt to learn what was going on in the community right now. 


Learn about the issues other operators are facing and determine areas we can work together. "Networking" in the people sense of the word. 


I am pls to spack wit San Francisco and not attend NANOG. 


I am here to have fun with my fellow friends in the community 


Drink beer 


Provide multicast feeds 


I run an R&E network.NANOG keeps me in touch with the "real world" problems and solutions. 


To know how to solve problems, it is interesting to meet vendors and discuss these issues. 


To understand the state of affairs on the internet and to learn as much as I can. 


NANOG is the only place I can avoid vendors and get real information without commercial spiels. 


The presentations on the state of BGP is directly relevant to PCH's research interests.We also used the opportunity to solicit input on current practices by core ISPs outside of the presentations themselves. 


Lets me have fun with other geeks, and reminds me why so much of my life is devoted to networking.:) 


As a customer of the providers, I wanted to learn more about the problems, solutions and ability of the group to work together. 


I come to NANOG in part to find out what others are doing, to figure out what my employer should be doing differently, in part to hear presentations on research that interests me, and in part to network with others I need to talk to to get specific things done. 


Have not attended recent meetings, wanted to hear current operational issues as well. 


Hear updates from ALL major exchange operators (i.e. not just PAIX). 


To meet with other operators which helps me in turn get more peers up. 



ISP Security



I learned some new techniques which I think are really valuable. 


good, good, good, really good 


Very informative, and very applicable given the state of the world. 


It is important for ISP's to share security techniques; it is good to see NANOG provide the vehicle. I hope it continues to happen at future meetings. 


The presentation seemed a bit disorganized. 


I enjoyed the real examples that were presented by the ISP 


While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 


Too much focus on router filtering features, and none on other aspects. For example, there could have been some coverage of host security, some of us do hosting too... 


The fact that we got to see the backscatter work in the 'lab' set up there was VERY helpful.Good show! 



While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 



IP Routing Protocol Scalability



The speaker tried to cover too much ground in the alloted time. A better focused detailed tutorial would be welcome. Please, this does not mean beginner level. 


It would be good to augment the topic with real-world examples to the extent posible 


Not very clear explanations at times. 


Learned a lot. will mostly profit from it because i learned which new knobs are in the protocols 


I found this session to be very informative. 


Spoke too fast, but otherwise a well-presented lecture. 


Generally very informative.Only negative comment would be having to listen to the many Cisco marketing plugs! 


While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


I was just torn between the two sessions.Since they were both continued after the break, I couldn't go to both.:( 


would like to see more technical depth. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 


This type of tutorial is almost always useful. 




The speaker tried to cover too much ground in the alloted time. A better focused detailed tutorial would be welcome. Please, this does not mean beginner level. 


While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


I think it could have been listed on the schedule as two halves; IGP, then BGP.If I'd known that going in, I would have done the security session first, then hopped across at the break for the BGP half of the scalability talk. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 



Designing a Testbed for Evaluating DDoS Defense Research


There was a lot of negativisim regarding reason why this could not be done rather than saying that it is needed and thinking out of the box on how to get over the current impedements that everyone kept harping on.A few good nuggests did rise to the surface though. 


The BoF produced some useful insights and information sharing between the people in the room. The best idea was the need to bring together people who do lab testing in order to share ideas and produce better labs (e.g., North American Test Network Operators Group - NATNOG ?) 


The motiviations for the testbed and the need for operator participation were not clearly articulated. 


No consensus could be reached. People in the crowd hijacked the meeting. Speaker had no control. The discussion quickly became circular. Overshot the time limit by a large margin and finally no result to go forward on. 


Good topic, bad problem :( tough audience. 


Purely a sales pitch for NAI labs, yuck. It wold have been more interesting to have the DDoS vendors or the secure Linux vendors give a talk. 



Data Center Peering (BOF IV)


We will see. 


A very useful function to have at a NANOG meeting, but I was disappointed to see many of the larger ISPs missing: AT&T, Level3, Verio, Qwest.net 


Needed more tech details. 


Can we have this one start at 8 pm next time?Too many folks were not able to get back from dinner meetings with various vendors/customers/peers in time for 7:30.Restaurants move slowly on Monday nights. 
The Peering BOF has been the most informative session in ages.It would be even better if there were an EIX-style reporting from NAP operators during the afternoon session on Monday. 


I think many new peerings agreements were started rolling here! 


Should be a regular feature. 


My favorite session in NANOG. 


Why do we need a BOF for peering coordinators to meet and give a word on why everyone should peer with their ISP?This lecture felt more like a half-party-group-half-get-together that didn't contain any information of useful value to anyone outside of directly needing to expand peering.A much more interesting topic which could've been discussed is peering policies and how they need to evolve to create a more stabalized global environment. 


That was a Riot:) 


The speaker enjoys having a big show and to sell himself The topic should be worked on more seriously. I probably will do that for Europe 


Bill's presentation was great as always. :> 


Very cool, where was Telstra?!!!!! 


Although a great opportunity to network and listen to other's strategies, however it would be even more beneficial to have a little more structure to the information sharing.Perhaps a written doc. compiled of everyone's strategies and objectives could be compiled for the attendees. 


Well-run and extremely useful. As a result of peering relationships initiated at this session, our company can reasonably expect to save in excess of $50k over the next year. This BOF alone paid for Earthlink's attendance. 


would like to see this done again.

"peering personals" was both entertaining and informative. 


The slides (or at least the ones identifying the participating peers should be posted on the web along with contact info for each participant. 


As usual, Bill Norton is great! 


Subject wasn't really of interest to me. 


Peering coordinator inroductions were very helpful, I will volunteer to introduce myself and our peering requirements at the next meeting. 


This was the most successful of the Peering BOF's or meeting I've been to. Very positive move to get the broad range of people giving descriptions of their networks.

Seems capital markets shrinking are opening peering policies and motivations.(need to lower expense) 


As always, Bill Norton is good, but there was some frustration--the number of people talking about their peering policies was fairly extensive, and trying to jot everything down so quickly led to hand cramps--a web page where the info would be made available afterwards would have been cool; even if there was just a page there saying "this will be filled in after the BOF" would have let us bookmark it for later reference...maybe next time around? 


It was an excellent venue to meet a variety of peering coordinators and get updates on their policies and locations. This is one of the strong suites of NANOG: the ability to "network" with the people who can help you expand your network.

Hat's off to Bill Norton for, though obviously a representative of Equinix, conducting an entirely neutral forum dedicated to BGP peering. 


Bill's peering forums continue to be extremely useful, educational, and entertaining.Having people stand up and say what they were looking for in peers is rather useful both in terms of figuring out who to peer with (something currently handled by somebody else in my company) and those of us with an academic interest in how interconnection between networks happens.

I understand including peering in the NANOG agenda is somewhat controversial. However, of all the areas of Internet operations, peering seems to be the one where who you know, and hotel bar conversations, is a huge determining factor in the amount and quality of peering it's possible to get.For that purpose, the NANOG conferences, or at least the gathering of the right people that tends to happen at NANOG conferences, is indispensable. 


Appropriate timing, of interest to audience, speaker keep audience engaged - evening meetings are difficult. 


I have been told that one of the people from Merit told Bill Norton that his presentation was "unprofessional" and that the little green stickers he put on people's cards wasn't a good idea.

If this is true, I *extremely* disagree. Frankly, I think the peering BOF was the most useful, informative, funny, laid back, and useful (yes, i'm saying it again) BOF the whole time I've ever gone to NANOG (my first one was NANOG 19).It also made people get together and talk with one another in a friendly fashion.

Please have another peering BOF with Bill Norton.It helped me out greatly. 


Bill did an excellent job of getting people together.It is often difficult to find the "right" people to talk to when arragning peering agreements. 



BGP Multihoming Guide


The speaker tried to cover too much ground in the alloted time. A better focused detailed tutorial would be welcome. Please, this does not mean beginner level. 


I learned a lot from this, and a lot of changed will hopefully be going into our network very soon. 


While I did not attend any of these sessions - I _really_ wanted to and plan on attending them at the next NANOG I attend.The topic choices were great. 


Very good basic summary of techniques, and presentation of less memory and CPU intensive designs for achieving effective link redundancy and load balancing. 


My lack of attendance shouldn't be seen as a strike against the tutorials. I skipped them because I felt they wouldn't teach me much I didn't already know, but a couple of years ago I would have gotten a lot out of them.I think those of us more experienced in this industry are able to make good use of that time in more informal discussions. 


Philip's description of BGP implementation was one of the most concise I've heard or read in awhile.He cut right to the point in his presentation.I've already passed his presentation onto a few of my peers to let it trickle down into thier organizations.EXCELLENT! Thanks. 






For Vendor Comments: 

Like hearing about what kinds of nets folks are building (ususally not our customers) and presentation on what kinds of features are needed, what techniques are used to operate today's networks. 


understand what's current security issues on the internet. will helping to test router 


To understand the developing needs of the ISP community. 


rekindle relationships (been out of networking last two years). 


Have not attended recent meetings, wanted to hear current operational issues as well. 



Hosting contact responses: 

Will contact you.

Greg Harp - [email protected]

Greg Hankins, Riverstone

[email protected]

[email protected]

[email protected]

[email protected]



Group Telecom is interested in hosting NANOG 25 (or other spring or fall) in Toronto, Canada.The person to contact is Randy Neals [email protected]
We have mentioned this to Susan at NANOG 21.CANARIE would help. I think we could do much better than Mtl.

[email protected]





^ Back to Top