presentation source
Download
Report
Transcript presentation source
The Direction of Value Flow in
Multi-service Connectionless Networks
Bob Briscoe
BT Research
7 Oct 1999
context
context
pricing
clearing
scenarios
summary
why think about this?
theory of the value of communications:
– connection-oriented hegemony
[ITU96, Zull97, Clark96, MacKie-Var92]
confusion is abroad in the world
– junk advertising, bundling content+network
why assume multicast sender pays?
alternative proposal seems obvious
– but wasn't when I started
7 Oct 1999
Direction of Value
3
context
context
pricing
clearing
scenarios
summary
context
multiservice, multicast IP (real-time apps)
some services usage-charged, whether
packet, flow or even longer term granularity
electronic tariff announcement
buyer's market
designed for bundling
– will cover in a future paper...
7 Oct 1999
Direction of Value
4
context
context
pricing
clearing
scenarios
summary
what business?
end-system
applic’n
present’n
link (wholesale cost)
network
connection? (not supplied)
session
e-commerce
transport
applic’n
present’n
session
transport
network
network router
network
network
link
link
link
link
physical
physical
physical
physical
7 Oct 1999
Direction of Value
5
context
context
pricing
clearing
scenarios
the value of place
value1 = f(i, place1 , time1 )
summary
i
value2 = f(i, place2 , time2 )
xmt value /leg = (value of place/time)
i
i
but competition cost+ margin pricing
who can know the value of place?
N: provider
Y: end-users
– apportionment
clearing grows comms market
7 Oct 1999
Direction of Value
6
context
pricing
pricing
clearing
scenarios
summary
end to end pricing
data flow
service
price
end-to-end
pricing role
7 Oct 1999
Direction of Value
9
context
pricing
pricing
clearing
scenarios
summary
end-to-end pricing role
not necessarily a business; a role taken by:
– one ISP
– the customer
– a third party (+ bundling content?)
but…
...what is the best default apportionment?...
7 Oct 1999
Direction of Value
11
context
pricing
pricing
clearing
scenarios
summary
default value apportionment
proposal:
– 'sender and receiver both charged'
– slight discrepancy in prices possible
– every leg of every flow is part charged twice
– model self-similar for inter-domain boundaries
7 Oct 1999
Direction of Value
12
context
pricing
pricing
clearing
scenarios
summary
split-edge pricing - per direction
Nd
Ne
Nd
price in & out
Nb
Wbas
Wabr
Na
Wbar
Wabs
Nc
Nc
7 Oct 1999
Direction of Value
separately
each price
between
boundary and
remote edge
price effects
localised
contracts
localised
extends
edge-pricing
[Shenker96]
13
context
pricing
pricing
clearing
split edge pricing
scenarios
-
summary
per class
each class
of service
has a
market
class
mismatch
effects
localised
7 Oct 1999
Direction of Value
14
context
pricing
pricing
clearing
scenarios
split-edge pricing
Nd
Qv
Ne
-
Wbas
Wabr
V Qt
7 Oct 1999
multicast &
Qw
Qw
Nb
Na
per leg
Nd
Qu
summary
heterogeneous
QoS
Qu
Wbar
Wabs
Qu
Nc
Nc
Direction of Value
15
context
pricing
pricing
clearing
scenarios
summary
default value apportionment
choose default to minimise clearing
'only senders' or 'only receivers' charged
halve transaction volume
require industry-wide co-operation
unstable - mavericks gain and increase costs of all
multicast requires trusted receiver count
'sender and receiver both charged'
majority of comms between consenting parties
value, charge and cost congruent
stable against maverick policies
doesn't match blame for transmission...
7 Oct 1999
Direction of Value
17
context
pricing
clearing
scenarios
summary
blame, liability and control
but…
...sender always to blame for a transmission
request-reply is irrelevant to network layer
other confusions clarified:
• value of moving info transitory
• value of moving GBs is often for a few bits
• cost of xmt << cost of targeting what to xmt
7 Oct 1999
Direction of Value
18
context
pricing
clearing
scenarios
summary
dilemma & solution
'sender and receiver both charged'
vs. 'sender to blame'
receiver customarily pays
...but...
sender liable
cf. postage stamp
• sender customarily pays, but receiver liable
7 Oct 1999
Direction of Value
19
context
pricing
clearing
scenarios
summary
clearing across edge providers
third party
service
7
money
3
2
5
46
44
5
51
49
clearing
7 Oct 1999
100
Direction of Value
21
context
pricing
clearing
scenarios
summary
clearing across edge providers
iterative
service
7
money
3
44
46
100
7 Oct 1999
54
44
51
Direction of Value
22
context
pricing
clearing
scenarios
summary
end-to-end model: justification
avoids Int'l Accounting Rate System flaws:
muddling wholesale with clearing
per-flow metering in backbone
trusted end counting for multicast
apportionment needs authentication
therefore unaggregateable
acd+1 prices in system 6e26
d5
a = apportionment
schemes per ISP
10
7 Oct 1999
Direction of Value
c 20,000 23
context
pricing
clearing
scenarios
summary
end-to-end model: justification II
localised contract
• jurisdiction clear
• what to meter decided locally
selective openness of clearing process
ac2 prices problem still exists 4e9
7 Oct 1999
Direction of Value
24
context
pricing
clearing
scenarios
scenarios
summary
scenarios
see paper for...
finding an end-to-end price
sender liable but local payment customary
inter-domain multicast with heterogeneous
quality of service (QoS)
phone to Internet gateway (PIG)
...ie. per session accounting
7 Oct 1999
Direction of Value
27
context
pricing
clearing
scenarios
scenarios
summary
PIG - phone to Internet gateways
PSTN
service
Internet
money
iterative
end-end
3
1
7
3
2
36
41
100
7 Oct 1999
5
21
5
clearing 59
29
25
24
21
PIG
Direction of Value
28
context
pricing
clearing
scenarios
summary
summary
alternative now seems obvious...
per-session accounting bill=anachronism
split-edge pricing scales infinitely
'sender and receiver both charged' stable
– customary - but only sender liable to pay
end-to-end pricing and clearing efficient
end-to-end intermediary could take 0-50%
of retail ISP and telco markets
– ...and their customer relationships
7 Oct 1999
Direction of Value
29
context
pricing
clearing
scenarios
more
summary
info
further information
Differential charging (diffchar) project
http://www.labs.bt.com/projects/mware/
Bob Briscoe
http://www.labs.bt.com/people/briscorj/
7 Oct 1999
Direction of Value
30