- Nsi-wg - lists.ogf.org

NSI demos for Supercomputing
by Jerry Sobieski 01 Nov '11

01 Nov '11

Hi All- I hope you all had a nice well deserved and relaxing weekend. Alas, we must get started on the final stretch this year and prepare for the Supercomputing demos... (there must be some cosmic justice to the fact that "demonstration" begins with "demons"...) The focus for SC will be to consolidate, stabilize, and demonstrate our existing capabilities. So, our target topology will be the topology we used last week for FIA, plus any added GOLEs or Networks that participate. There are two ways to participate: a) as an NSI network running an NSA, or b) as a supporting organization that is willing to lend transport resources between these NSI networks or is willing to help write software tools, etc. I will acknowledge all organizations on the posters/handouts. ---> I want to freeze the topology by November 7th so that applications building off the diagrams don't have to keep fixing their graphics everytime someone realizes they want to participate...So if you want to be part of this, let me know ASAP. (That means yesterday!:-)) The fundamental objective of the SC demo is to set up circuits using NSI-CS across multiple domains and over real hardware. So each network will need to verify their 1) control plane interoperability with the other NSAs, and 2) insure that their data plane is properly configured between your network and your adjacent peer. Attached is the initial topology diagram and the associated .owl file for SC - SC2011-Topo-v1.0 Changes from latest FIA topo: - Added STPs for USLHCnet between NL and SL. For SC, we will again use only four VLANs on each link: 1780 to 1783. In some locations we may need to only use a subset of these depending upon how the NSAs are resolved by next week. Tasks: For the GOLE/network operators: pursue the following tasks: 1) Every network should have at least one end system that can be used to terminate circuits and that can issue and respond to IP layer Pings. This could be a perfSonar host with apropriate access provided to the demo team, or a small Linux or Mac, or possibly the switch processor itself. But we need something we can use to test to, and access to configure and verify the circuit. 2) Control plane: Begin testing immediately from your NSA to your neighbor's NSA to set up simple intra-domain Reservation Requests across your neighbor's network, and then progress to test inter-domain reservations across your network *and* your neighbor. THis is to insure proper and reliable interoperability between NSAs. Coordinate with your neighbor network (you all know the PoCs by now) to test. 3) Data plane: Ping test the connections to your neighbors' networks to insure the VLANs are clear channels between the NSI Networks. The crossconnects within your networks will never work if the static VLANs between networks are not operational first. So _/verify these VLANs/_ by manually placing an end system from each network on each VLAN, configuring the IP subnet, and pinging one another. For the code developers: 1) Make sure your hardware/NRM interfaces are working reliably. 2) Make sure you are able to deal with SOAP/HTTPS messaging. If you need to implement it - now is the time. If there are issues doing so, lets get those exposed so that we can resolve them. Monitor the nsi-wg chat on Skype... 3) Make sure the Query command is working properly. 4) Lets get the Query Agent, the MotherEarth (autoEarth) vizualizer, and the AIST-vizualizer working *RELIABLY*. You all should dive into this as soon as the SC WSDL interop, query interop, and HTTPS is working. This is the capstone task for this year!!! These visualizers will exercise and demonstrate all of the other great work we've accomplished. So let make these look good! Thanks all.... Any additions, corrections, thoughts, or comments, please post or send them to me and we'll integrate them into the task tracking... Two weeks till Showtime! Jerry

2 1

Interoperability matrix for SC
by Takahiro Miyamoto 31 Oct '11

31 Oct '11

Hi all, I made a matrix sheet for interoperability of WSDL 1.SC instead of the matrix for 1.Rio. https://docs.google.com/spreadsheet/ccc?key=0AkOr3QHUMxp-dFB6TDRtajNQYUl3bm… Please update your status. Best regards, Takahiro -- Takahiro Miyamoto <tk-miyamoto(a)kddilabs.jp> KDDI R&D Laboratories, Inc. TEL: +81-49-278-7455 / FAX: +81-49-278-7811

1 0

NSI demonstration at FIA in Poznan
by Radek Krzywania 27 Oct '11

27 Oct '11

Hi all, We have a working successful presentation of NSI with data plane enabled. We are able to setup a global circuit between PIONIER and AIST networks in automated manner. The control plane is successfully visualised on Automated Earth interface and a data plane is verified with ICMP messages (PING). Thanks to involvement of many individuals and organisations, we have made a big step in the direction of SC'11 demo and further NSI development. I would especially thank Michał Balcerkiewicz (PSNC), Jeroen van der Ham (UvA), Takatoshi Ikeda (AIST), Tomohiro Kudoh (AIST), Henrik Thostrup Jensen (NORDUNET), Gerben van Malenstein (SURFNET), John MacAuley (SURFNET) and Jerry Sobieski (NORDUNET) for sacrificing their time and did the best to make it working (apology if I missed here anyone). The NSI booth and a demo was visited by multiple conference participants, including representatives of Brasilian NREN RNP, I2CAT, University of Essex, GEANT, GRNET/NTUA, NextWorks and more. Once again, thank you all NSI folks for helping with this event and demo preparation and execution! Best regards Radek ________________________________________________________________________ Radoslaw Krzywania Network Research and Development Poznan Supercomputing and radek.krzywania(a)man.poznan.pl Networking Center +48 61 850 25 26 http://www.man.poznan.pl ________________________________________________________________________

2 1

FIA-Topo-v1.4d
by Jerry Sobieski 27 Oct '11

27 Oct '11

Hi all- Here is version 1.4d topofile. Changes: - set up vlan 1781 to be statically configured thru NetherLight from Pionier to StarLight. (commented out the 1781 STPs in NL on the pionier port and the SL port, and set the remote peer in the ports at SL and Pioner. - updated the csProviderEndpoint for pionier. - modified the Location object for GEANT to point at Paris rather than Cambridge. Thanks Jerry

3 2

Tomorrow's NSI call - please respond.
by Inder Monga 26 Oct '11

26 Oct '11

Hi All, Please let the list know if you can make the call tomorrow and if you are prepared to talk about Security. This is an important topic and I would love to have as close to full attendance as possible so we do not have to re-do the discussion. Your response is appreciated, Inder -- Inder Monga 510-486-6531 http://www.es.net Follow us on Twitter: ESnetUpdates/Twitter <http://bit.ly/bisCAd> Visit our blog: ESnetUpdates Blog <http://bit.ly/9lSTO3><http://bit.ly/d2Olql>

3 3

Re: [Nsi-wg] Issue 28 in ogf-nsi-project: Security
by ogf-nsi-project＠googlecode.com 25 Oct '11

25 Oct '11

Comment #5 on issue 28 by thost...(a)gmail.com: Security http://code.google.com/p/ogf-nsi-project/issues/detail?id=28 I'd like to try and get SAML out of the WSDL before the SC11 demo. SAML is typically used in single sign on between identity provider and service providers. It does not really fit into NSI. We may have the need to have something like sessionSecurityAttr, though often these attributes are not related to security, but are typically request meta-data. Perhaps we could just call them sessionAttributes. Anyay I think the scheme with Name/Value in the protocol spec. e.g.: <sessionAttribute> <name> ... </name> <value> ... </value> </sessionAttribute> and a slighly more real example: <sessionAttribute> <name>requesterIdentity</name> <value>nsiuser(a)example.com</value> </sessionAttribute> The idea here is to have a place for connection information. What it should NOT be used for is any authentication or authorization, and ideally the attributs can be stripped without any reprocussions. If possible I would like to use this for SC11 visualization query tool, which could add something like: <sessionAttribute> <name>requesterIdentity</name> <value>NSIVisualizationTool</value> </sessionAttribute> Which would indicate to the NSI agent, that it should return all connections in the query made. This can of course be combined with proper authentication/authorization when we have that, but right we don't. Well there is the IP address, which can be made available from the MTL / protocol layer.

1 5

FIA-Topo-v1.4c.owl
by Jerry Sobieski 25 Oct '11

25 Oct '11

Changes: - corrected urn.ogf.network to be urn:ogf:network... J

1 0

FIA topo 1.4b
by Jerry Sobieski 25 Oct '11

25 Oct '11

- Changed comment markers to correctly comment out netherlight stps for 1780. Jerry

1 0

FIA Topo v1.4a
by Jerry Sobieski 25 Oct '11

25 Oct '11

Hi all - yet another topo mod.. - Changed the interfaces at StarLight and Amsterdam to reflect that valn 1780 was no longer available on the AMS-CHI IRNc link... The new pdf shows the mapping now. Jerry

1 0

FIA-Topo-v1.4.owl
by Jerry Sobieski 25 Oct '11

25 Oct '11

New topo file Changes: - modified pionier.ets:ams-80 to be SDP with aist.ets:chi-80 i.e. vlan 1780 is statically configured thru CHI and AMS so Pionier and AISt are adjacent on that VLAN. Jerry

1 0