September 2010 - Nsi-wg - lists.ogf.org

time issue
by John Vollbrecht 07 Oct '10

07 Oct '10

Hello all - Jerry and I had a discussion last week about the time issue. I think we developed a useful approach. The idea is to define two times, which I think we all agree exist. 1) available time - time a connection is available to the application to communicate between devices 2) resource time - time a resource is reserved to support available time To further define these - Available time - requested by the user for its application - provided by the network. resource time - time a resource is allocated to a connection - includes setup and teardown time, if any - is time in reservation calendar for resource Available time requested cannot be provided exactly by network because it cannot predict exactly length of setup and take down. I believe we all agree with this. Therefore provided available time can at best approximate requested available time. We agreed that when a user requests automatic start connection it would request available time and the provider would schedule resource time to get as close as possible. When a request is for user initiated connection the time would be for reserved time, and the user initiation can start anytime after the reserved time. Available time depends on setup and take down times of equipment. ----------------- I think we agreed on the above definitions. The definition of time seem useful in discussing what goes in connection service messages. We also talked about some possible implications of this. The difference between available and resource time is setup and takedown time. While it is impossible to be sure exactly how long they will be, it may be possible to define something statistical. For example setup takes an average of 17 sec with std deviation of N. If this is can be defined for the resource, then one can make a prediction about when a connection will be available with a degree of confidence. For example this would allow one to request an automatic connection, for example, at 5pm and have it available 99% of the time. If the average setup time is 17 seconds and I add 10 seconds to be 99% sure, then the service would initiate the connection at 5:00:00 - 0:00:25, or 4:59:35. We talked about including this "setup requirement" in the connection service definition of and NSA, and by implication including this in requests and replies. I think this is worth talking about in the group. John

11 48

Connection services in ION and ESnet..
by Jerry Sobieski 01 Oct '10

01 Oct '10

Hi all- In our GN3 SA2-T1 wg regarding the Bandwdith on Demand service, we were discussing bandwidth granularity requirements. Could folks from ESnet and I2 comment on what you think appropriate granularity should be for these types of connection services and why? In the SA2-T1 group, applications such as FTPs, videoconferencing, etc were brought up - indicating a wide range of possible bandwidth requirements... Any thoughts or experiences would be appreciated. Jerry

2 1

Call problem
by John Vollbrecht 29 Sep '10

29 Sep '10

I am not sure if anyone tried to call this morning, but there is a problem with the call Apparently the conference call terminated and I can't restore the same id. I have set up the following call. If anyone wants to call in I will be there. The call usually starts at 10:30 - jrv(a)internet2.edu has invited you to a conference call. Starting Time: Sep 29, 2010 at 9:30 AM, US/Eastern Call: +1-734-615-7474 (Please use if you do not pay for Long Distance) or +1-866-411-0013 (toll free US/Canada Only) Enter access code: 0124096

1 0

minutes from today's NSI call
by Guy Roberts 23 Sep '10

23 Sep '10

... are available here: http://forge.gridforum.org/sf/go/doc16061?nav=1 Guy

1 0

NSI advance reservation parameters
by Guy Roberts 23 Sep '10

23 Sep '10

To help along the discussion on advance reservation I include a list of decision previously made by the group... Guy Issue v1.0 Comments 1 or 2 phase commit? 1PC 2PC is considered unnecessary for v1.0 Is automatic provisioning (initiated by timer) required? Yes Is explicit provisioning (initiated by requester message) required? Yes Is advance reservation required? Yes Is immediate reservation required? Yes May need to update architecture text Allow ASAP as start-time for immediate reservation? Yes How is advance distinguished from immediate reservation? Immediate uses 'ASAP' in the start-time field What does start-time mean as used in protocol for auto mode? In-service begin time It is the responsibility of the provider NSA to make sure that this in-service start time is met What does start-time mean as used in protocol for explicit mode? Time at which the provider is able to accept a provision signal It is now the requestor's responsibility to advance the explicit signal to ensure good in-service time. What does end-time mean as used in protocol for explicit mode? End of reservation What does end-time mean as used in protocol for advance reservation? In-service end time. This is the same as beginning of de-provisioning time Is a tear-down signal required? Yes Is connection duration required? May be required for ASAP start time Duration would replace end time. Is unspecified end-time allowed? No agreement on this point Provider validation of start time ? Validated This is a constraint in path finding Is a flag necessary to indicate immediate vs. advance reservation No Not necessary, immediate is indicated with 'ASAP' start-time. See flag below. Is a flag needed to indicate whether start-time later than specified guard is allowed? Yes These start times are treated as 'now' and warning may be sent back to the Requester NSA. Does each Connection request need a flag to indicate how to expect provisioning to be initiated? Yes, flag indicating provisioning method is required Does a reservation request require a pass/fail response from the Provider NSA? Response sent once reservation process completed. Notification of provisioning? Notification of provisioning completion is required.

1 0

FW: Today's call
by Guy Roberts 22 Sep '10

22 Sep '10

Please find attached presentation from Tomohiro Kudoh regarding agenda item 3 for today's NSI conf call. Guy -----Original Message----- From: Tomohiro Kudoh [mailto:t.kudoh@aist.go.jp] Sent: 22 September 2010 13:33 To: Guy Roberts Subject: Today's call Hello Guy, I am now attending ECOC conference at Torino, and will not be able to participate in today's call. By the way, the slide I sent to the list on Apr.21 (attached) might be related to the start time and end time issue currently being discussed at the call. Tomohiro

1 0

today's NSI call
by Guy Roberts 22 Sep '10

22 Sep '10

Hi all, The following is dial-in information for Wednesday's NSI call: Time: 7:30 PDT 10:30 EDT, 15:30 BST, 16:30 CEST, 23:30 JST Call: +1-734-615-7474 (Please use if you do not pay for Long Distance) or +1-866-411-0013 (toll free US/Canada Only) Enter access code: 0155180 Thanks to John Vollbrecht for the agenda: 1) Do we agree that Modify primitive for Connections will not be in V.1 - I agree, though I think it could (maybe should) be in the architecture V1 doc and not in the protocol V1. 2) Do we agree that Notify will be in V1 - I agree, the question is whether Notify is a service (for NSA as a whole) or a primitive (for connection service) or both. I vote for it being a service but could be either and changed in public comment time if we so decide. 3) How do we handle start and end time? - this is a trickier issue. Seems to me there are two users of time - the application using the network, and the network allocating resources to the connection. Whatever we decide needs to be usable by both classes of users. - for applications they need to know when the connection is available for them. -an argument is that the network should send a notify to the user when the network is "available" . My take is that this is good, and many applications will use it. regardless of whether a notify is sent, one must decide whether time in a request is for available time, or time to start making the connection. I am not sure there is a good answer unless one were to somehow include both, which has not been discussed before. -------------------------- A new issue: The proposed connection service has two fields (see sec 3 in the attached doc), that describe the connection. One field, the serviceInstance-attribute describes the performance characteristics of the connection, and is use by the application to define the characteristics it needs in the connection. The other field describes the topology of the connection and is used by network providers to reserve and provision the network. The issue I think worth discussing is what is in the topology or path field. My thinking is is that in the confirm message at least, the path field has information about the connection itself and any sub-connections. It contains the ingress of the network and potentially segments in between. Segments may not be included for several reasons, including that networks may not want their information exported beyond the original requestor, or because there are no subnetworks. A path field then would contain something like [NSA1:CID-1, Ingress1[ [NSAa:CID-a, Ingress-1,Egress-a] [NSAb:CID-b, Ingress-b, Egress-b] [NSAc:CID-c, Ingress-c,Egress-1], Egress1] This is not well defined, the intent is that it MUST contain CID, Ingress and Egress. It MAY contain segments that are connections in their own right. This could be nested as well. The CIDs include the id of the NSA that authorized the connection. One could imagine other elements in this field, like time. This might be used to carry network reservation time, while the service attribute would carry available time. This is something to think about.

1 0

Issues for discussion
by John Vollbrecht 22 Sep '10

22 Sep '10

A set of issues that we might discuss tomorrow. Some we have talked about before: 1) Do we agree that Modify primitive for Connections will not be in V.1 - I agree, though I think it could (maybe should) be in the architecture V1 doc and not in the protocol V1. 2) Do we agree that Notify will be in V1 - I agree, the question is whether Notify is a service (for NSA as a whole) or a primitive (for connection service) or both. I vote for it being a service but could be either and changed in public comment time if we so decide. 3) How do we handle start and end time? - this is a trickier issue. Seems to me there are two users of time - the application using the network, and the network allocating resources to the connection. Whatever we decide needs to be usable by both classes of users. - for applications they need to know when the connection is available for them. -an argument is that the network should send a notify to the user when the network is "available" . My take is that this is good, and many applications will use it. regardless of whether a notify is sent, one must decide whether time in a request is for available time, or time to start making the connection. I am not sure there is a good answer unless one were to somehow include both, which has not been discussed before. -------------------------- A new issue: The proposed connection service has two fields (see sec 3 in the attached doc), that describe the connection. One field, the serviceInstance-attribute describes the performance characteristics of the connection, and is use by the application to define the characteristics it needs in the connection. The other field describes the topology of the connection and is used by network providers to reserve and provision the network. The issue I think worth discussing is what is in the topology or path field. My thinking is is that in the confirm message at least, the path field has information about the connection itself and any sub-connections. It contains the ingress of the network and potentially segments in between. Segments may not be included for several reasons, including that networks may not want their information exported beyond the original requestor, or because there are no subnetworks. A path field then would contain something like [NSA1:CID-1, Ingress1[ [NSAa:CID-a, Ingress-1,Egress-a] [NSAb:CID-b, Ingress-b, Egress-b] [NSAc:CID-c, Ingress-c,Egress-1], Egress1] This is not well defined, the intent is that it MUST contain CID, Ingress and Egress. It MAY contain segments that are connections in their own right. This could be nested as well. The CIDs include the id of the NSA that authorized the connection. One could imagine other elements in this field, like time. This might be used to carry network reservation time, while the service attribute would carry available time. This is something to think about.

1 0

ANI Testbed for Network Research: Call for proposals
by Inder Monga 22 Sep '10

22 Sep '10

Attached is a call for proposals for ESnet's Network Research Testbed. Please feel free to submit a proposal and forward it to researchers who might be able to take advantage of this opportunity. Thanks, Inder --- ESnet Call for Proposals: Advanced Network Initiative Testbed ESnet is now soliciting research proposals for its ARRA-funded testbed. It currently provides network researchers with a rapidly reconfigurable high-performance network research environment where reproducible tests can be run. This will eventually evolve into a nationwide 100Gbps testbed, available for use to any researcher whose proposal is accepted. Sample Research Researchers can use the testbed to prototype, test, and validate cutting edge networking concepts, for example, projects including: • Path computation algorithms that incorporate information about hybrid layer 1, 2 and 3 paths, and support 'cut-through' routing. • New transport protocols for high speed networks • Protection and recovery algorithms • Automatic classification of large bulk data flows • New routing protocols • New network management techniques • Novel packet processing algorithms • High-throughput middleware and applications research Please look at the description https://sites.google.com/a/lbl.gov/ani-testbed/testbed-description to get a more detailed idea of the current testbed capabilities. Important Dates The proposal review panel will discuss and review proposals twice yearly. The first round of proposals is due October 1, 2010, and decisions will be made by Dec 10, 2011. After that the committee will meet approximately every six months to accept additional proposals and review progress of current projects. Proposals should be sent to: ani-testbed-proposal(a)es.net More details on the testbed and the brief proposal process can be found here: https://sites.google.com/a/lbl.gov/ani-testbed How to Apply Researchers wanting to access the ANI testbed must submit a short 1-2 page PDF proposal, including detailed information on: • Proposed research • Testbed usage o Testbed components used o Time and access requirements For further proposal process and requirements details, please visit: https://sites.google.com/a/lbl.gov/ani-testbed/proposal-process About the Advanced Network Initiative As part of the ARRA Advanced Networking Initiative, launched by $62 million in American Recovery and Reinvestment Act funds, ESnet www.es.net has embarked on a project to handle the expanding data needs between DOE supercomputing facilities. The goal of the project is to accelerate deployment of 100 Gbps technologies to support science research and education, as well as build a robust infrastructure that will transition into a production network. The testbed will be located in Lawrence Berkeley National Labs at ESnet facilities until October 2010. The testbed will move to Long Island on a dark fiber ring between Brookhaven National Lab (BNL) and Manhattan (MANLAN) by the end of 2010. In the future, new capabilities may be added to the location of the current testbed. --- Inder Monga ANI Testbed imonga(a)es.net ESnet Blog (510) 499 8065 (c) (510) 486 6531 (o) "Whatever your mind can conceive and believe it can achieve." - Napoleon Hill

1 0

NSI sessions requested at OGF30
by Guy Roberts 17 Sep '10

17 Sep '10

Hello All, I have requested 3 NSI sessions at OGF30, for details view these links (note you may need gridforge account for this) http://www.ogf.org/gf/session_request/editreq.php?id=2109&key=a45345ef http://www.ogf.org/gf/session_request/editreq.php?id=2110&key=8902e28b http://www.ogf.org/gf/session_request/editreq.php?id=2111&key=b0b0fca5 Guy

1 0