kwe@bu-cs.BU.EDU (kwe@bu-it.bu.edu (Kent W. England)) (04/29/88)
All this talk about problems and issues that would happen if cost recovery just starts up ad-hoc proves that a straightforward attempt to implement cost recovery piecemeal will fail. I would think that one place to start in coming up with something reasonable for a campus/regional/backbone hierarchical model would be to look at the network management data in the Management Information Base (MIB). IpPktsIn, IpPktsOut, TcpPktsIn, etc. I don't think usage charges should get any more complicated than that for LAN/internet traffic. If the jvnc-net people sent me a bill for $xxx per month line charge for my 56kbps link and $yyy per thousand IpPktsIn/IpPktsOut I could check that against what my router tells me went in/out that interface. I could set up the same kind of billing per k of IpPkts for my local Ethernets if I wanted. I really don't think it should get any more complex than that. Even this simple scheme has problems. What if BITnet and jvncnet have vastly different charging schemes? What if jvncnet and the nsfnet backbone have different schemes? What if jvncnet wants to bill me for traffic that goes onto the backbone? How will they measure that? How will I verify or understand charges like that? There have to be simple rules for collecting traffic information and collecting charges at limited specific points in the network. I don't see how it will work unless the campus nets, the regionals, and the backbones all have similar models for collecting money. Services like the backbone should charge the regionals only and the regionals pass on charges to the campus LANs without trying to account for individual packets as they traverse nets. Trying to differentiate intra-regional from inter-regional (sound like intraLATA vs interLATA? :-) would be too complicated to do at first. Point is, it has to be coordinated among the various nets or we have games being played where traffic seeks the least literal cost route and not most effective transport route. Keep it simple. Kent England, Boston University
steve@NOTE.NSF.GOV (Stephen Wolff) (04/29/88)
The NSFNET backbone will be directly funded by NSF. -s
heker@JVNCA.CSC.ORG (Sergio Heker) (04/29/88)
Kent, You have made two valid points here that should be differenciated, one is how to charge (fix cost versus measured traffic cost), second is consistency among all the mid-level networks (i.e. JVNCNet) and the backbone networks (i.e. NSFNET). On the first point you have to cover your costs whether you charge for measured traffic or fix cost. The problem with measured traffic is that: (1) we have to further agree among all the other networks administrations what parameters to use in the charging model, (2) the implementation can become "expensive" to accomplish from the bandwidth and cpu point of view. Thus I suggest the fix cost depending on line bandwidth (as I recall Craig Partridge suggested this before). These costs have to: (1) cover the operational costs of the network, and (2) be consistent among other networks. All this can be worked on, with the exception that this assumes one type of user (i.e. University), what about if the user of the network is a researcher working for a profit making organization?, shall he pay the same as a researcher working for a University?. The networks then have to get into an aggreement that is consistent, and take all these issues in consideration. As a point of information, there is a task force whithin the Federation of American Research Networks (FARnet), working on this issue right now. -- Sergio ----------------------------------------------------------------------------- Sergio Heker tel: (609) 520-2000 Internet: "heker@jvnca.csc.org" Bitnet: "heker@jvnc" JOHN VON NEUMANN NATIONAL SUPERCOMPUTER CENTER, JVNCnet Network Manager -----------------------------------------------------------------------------
bernhold@qtp.ufl.edu (David E. Bernholdt) (05/02/88)
In article <8804291239.AA21630@jvnca.csc.org> heker@JVNCA.CSC.ORG (Sergio Heker) writes: >Thus I suggest >the fix cost depending on line bandwidth (as I recall Craig Partridge >suggested this before). It seems to me that in the long run this is going to be a problem. If sites are charged based on the bandwidth of their connection to the net, most bean-counters aren't likely to be overly generous with the capacity they are willing to fund. Then nobody is going to have the bandwidth for any more than their own usage & nobody will be willing to sacrifice their much-needed bandwidth to pass somebody else's packets on to another destination. And very quickly the net looses its usefulness. Is my view overly simplistic or too pessimistic (sp?)?? Dave -- David Bernholdt Internet: bernhold@orange.qtp.ufl.edu Quantum Theory Project BITnet: bernhold@ufpine.bitnet University of Florida HEPnet: 43129::59410::bernholdt Gainesville, FL 32611 Phone: 904/392 9306