Participants: Dale Finkelson, Paul Schopis, Michael Lambert, Russ Hobby, Jeff Bartig, Brian Cashman, Steve Wallace, Chris Robb, Matt Davy, Jon-Paul Herron, Joe St. Sauver, Ana Preston, Andrea Blome, John Moore, Rob Vietzke, Tom Lehman, Linda Winkler, Eric Boyd, Rick Summerhill, Bill Owens, Matt Zekauskas, James Deaton, Cas D'Angelo, Dave Reese, Dave Farmer
Agenda
• Update from Internet2 (Chris Robb) - 5 min
o Things have been fairly quiet. Making progress on Charlotte colo build with commit date of 5/31/08 from Level 3 for site handover on the colocation suite, power, Infinera node augmentation. Looking at installing Ciena node in early June. Transition strategy for inserting node will be hot cut. Sitting on old Infinera 3.0 code. Infinera released code fix 4.01 that solved one problem but introduced more. We have backed off of that upgrade and waiting for new code. We are ready to go with upgrade within the next 2 weeks. Looking to have everything up by October 1st. Installing gear in early August. Have an additional T640 that can be used for migration.
• Update from AOAC (Paul Schopis) - 5 Min
o Strategic documents and governance model were reviewed at the face-to-face meeting yesterday.
o Conflict resolution using the Internet2 policy - AOAC is the body that is called to fix those issues. Still finding out what the council is charged with - does the council make policy or create policy recommendations for the Board? Looking at business models and potential service offerings. Feel free to contact any AOAC member regarding anything you feel may fall under the AOAC jurisdiction.
o There is a new RHCPP working group that is a joint NTAC /AOAC project to review both policy and technical aspect of the issue. The RHCPP working group's purpose is to resolve issues such as what is happening in Oregon. Their rural health pilot has a number of small ISPs that provide services to small end sites and they don't have equipment or expertise to assist in the connection. Dave Crowe is the chair of this group.
• Update from Peering Working Group including Quarterly Peering Report (Jeff Bartig) - 10 min
o Quarterly update on Internet2 for CPS:
• Migration to v6 is almost completed. The deadline is the end of this month to complete and there are four aren't completed. As for the IPv4 peering, there are nearly 9000 prefixes, 54 peers and roughly 1000 of our prefixes to those peers. There are 24 connector peering sessions and 6 new peers this quarter. Still experiencing issues gathering traffic statistics. The atlas problem has been fixed, so weather map is operational. The goal is to have the minimum set of tools to get started. Tools are almost finished. Things are very manual at this time. Some additional tools will be created and are looking at the route population as the target audience. Some of the data is available today, but no aggregates at this time. We will look into it and get back to the group.
• 5. Report from RHCPP Working Group (David Crowe) - 5 min
o See above
• 6. Report from DCN Working Groups (Eric Boyd) - 5 min
o DCN connectors:
• CENIC
• MAX
• Merit
• NoX
• NYSERNet
• PNWGP
• Five others in discussion
o DCN virtual organization participation:
• LHC Atlas
• LHC CMS
• LIGO
• Two others in discussion
o DCN applications:
• Terapaths
• LambdaStation
• Phoebus
• Three others in discussion
o DCN software suite:
• OSCARS (IDC)
• DRAGON (DC)
• Version 0.3 of DCNSS released April 18, 2008; version 0.4 to be released Summer 2008
o Formal DCN working group met this morning (operations group). Sign-up sheets from all of the DCN related sessions were distributed. It was decided to make one mailing list for the working group. The group will begin with attendees from this week's sessions and then reach out to connectors if they are not already on the list. There is no requirement to be a member of the NTAC to participate in any working group.
o Need to find a way to advertise that there are no restrictions to be a member of a working group. Perhaps the Internet2 newsletter?
• 7. Report from Routing Policy Working Group (David Farmer) - 5 min
o The working group was just formed at the beginning of the monthin response to the Akamai situation. The group is refining the scope of their activities and would appreciate input from the NTAC. Fundamentally will be addressing issues regarding announced routes from Internet2 and R&E to participants without consulting with the community. Should we talk about general routing policies and issues, i.e., why doesn't xxx prefer the Internet2 routes over other routes? Perhaps setting some guidelines would be helpful; announce at least specific routes to Internet2 instead of the global route table.
o What kind of view should we have for this working group?
• Recipes on the NOC pages were useful but need updating.
• v6 aggregation
o Suggestions:
• Sharing of best practices
• Aggregation to v6
• Common practice of how things are done in the real world.
• Would like to see the group tackle broader scale issues.
• Would be nice to put something out there for people to refer to, i.e., policy and documents.
• Would like to discuss multi-task issue, in terms of best common practices. There are no good places to find that information unless you've been a part of this community for some time (BCP). There is a multi-task working group - should take this to them. Linda will seek them out.
• 8. Update on ARIN issues (Dale Finkelson) - 5 min
o John Kern speaking at SMM regarding v4 address exhaustion on ARIN. The working group decided that these issues are important to the entire Internet2 membership and not just those attending Joint Techs.
o Current hot topic: Transfer policies; what should it look like, should we have one? Please send Dale Finkelson any opinions you may have.
o Do folks think we should ask membership about legacy?
• ARIN took away an action item from last meeting to clean up their agreement about this.
• ARIN will tweak agreement for state agencies and state entities, so they recognize they have to do that. Please send any specific issues ASAP.
• Should Internet2 be making a recommendation one way or another to the membership? No
• What are the sections of the agreement that folks have issues with?
• Ability to change the agreement;
• Sent this information to Dale and he will formalize something and take to ARIN. This needs to be done ASAP.
• What are the possibilities of modifying the agreement? Current statement identifies your right to use these resources in a legally defined way. As we come to exhaustion we are going to need this well defined.
• 7. Update on naval observatory (David Farmer/Jose Domingues) - 5 min
o N/A
• 8. other topics - 30 min
o New working group descriptions were sent out. This is just a starting point. Let the group know if you are interested in helping get these going. This came about because of discussions regarding ITECS and at one point there was a lot more community involvement in engineering and testing. The working groups represent areas that the community can drive the efforts in. Where does Internet2 need help from the community? Internet2 staff sees a need for the working group.
o Measurement Working Group (leaderless): Set of tools out there. There are a lot of other tools that need to be developed. Need to identify what kinds of tools should be developed. Ongoing work on framework space for perfSONAR. Even if we have software available, the community needs to decide what they want to do.
• Is the working group similar to the former measurement sig? Ongoing sig like environment, but the community hasn't said where they want it to go. Would be valuable to find out where we are going and how to deploy. Out of the realm of a sig into a more action oriented working group.
• Should be a very focused, wider ranged group. Don't know what the charter should be. Valuable to have outside folks focusing on where to go, what's missing, etc.
• RONs would like to have accessible documents on measurements.
o Phoebus working group (named as place holder for now -- leaderless):
• Think of new name.
• Intention is to encourage people to use Phoebus.
• Internet2 is making an effort to deploy Phoebus - no guidance needed.
• Martin Swaney and his students are developing this. He is happy to have input from those who have used it.
• Need to set a project plan path so that folks don't come back and say we shouldn't invest. This would be helpful for the group to investigate, including connection agreements, etc.
• Need to rethink assumptions.
• This group should develop project plan for all projects, not just Phoebus.
• Need to look at the higher level stuff.
o Information and topology group (leaderless)