Telecommunications @ Crossroads
Philip K. Mutooni
President
IP.Phusion Technologies, Inc.
1
Current Network Infrastructure
RBOC Toll Switch
RBOC Toll Switch
IXC Network
Local
Exchange
Local
Exchange
IP Network
Local ISP
Local ISP
2
The Traffic Transition Model
■
■
Assumptions:
–
Installed Capacity closely matched with peak traffic
–
Capacity growing at constant compounded rate
Formulation:
–
C(τ)=Co(1+rx)τ
where Co is a base capacity, rx is a voice or data growth rate, and τ is a time duration
–
Cov=αCod
where α is a constant determined during the model’s calibration
3
Model Formulation
GROWTH OF DATA AND VOICE CAPACITIES
900.00
Relative Capacity
800.00
700.00
600.00
500.00
Voice Traffic
400.00
Data Traffic
300.00
200.00
10
9
8
7
6
tc
5
4
3
2
to
0
Cod
0.00
1
Cov
100.00
Time
* The general function used to plot these graphs was k(1+x)t. A different value of x was used in
both the voice and data cases with x(voice) < x(data).
4
Desired Model Results
■
Five quantities of interest:
1
tl or the lead-user point:
Defn - The point at which packet-data traffic required 10% of the total capacity
2
tc or the crossover point:
Defn - The point where both types of service require the same capacity
3
te or the eclipse point:
Defn - The point where packet-data traffic consumes 90% of overall backbone capacity
4
5
The time interval tc-tl
The time interval te-tc
5
Calibrating the Traffic Transition Model
■
Four Step Process:
1
2
3
4
■
Obtain Overall Capacity Ctotal at known points
Decompose Ctotal into Cvoice and Cdata components
Determine growth rates rv, rd, and rtotal
Determine tl, tc, te
Process applied to two data sets:
–
–
For a selected MCI POP
For a selected ATT POP
6
Model Results - MCI Case
8.5
8
7.5
7
6.5
6
5.5
5
4.5
4
3.5
3
2.5
2
1.5
1
Log Voice Capacity
Log Data Capacity
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
Log Total Capacity
1988
Log10(Capacity)
MCI Traffic Transition Results (1988-2010)
Year
tl
tc
te
t c- t l
te-tc
1991
Jan 1 9 9 9
2007
8 yrs
8 yrs
7
Model Results - ATT Case
AT&T Traffic Transition Results (1988-2010)
5.5
5
4.5
4
3.5
Log Voice Capacity
3
Log Data Capacity
2.5
2010
2008
2006
2002
2000
1998
1996
1994
1992
1990
2004
Log Total Capacity
2
1988
Log10(Capacity)
6
Year
tl
tc
te
tc-tl
te-tc
1988
A pr 1997
2006
9 yrs
9 yrs
8
Model Results - Industry-wide
8.5
8
7.5
7
6.5
6
5.5
5
4.5
4
3.5
3
2.5
2
Log Voice Capacity
Log Data Capacity
2010
2008
2006
2004
Year
2002
2000
1998
1996
1994
1992
1990
Log Total Capacity
1988
Log10(Capacity)
Industry-Wide Traffic Transition Results
(1988-2010)
tl
tc
te
t c -t l
t e -t c
1988
N o v 1998
2007
10 years
9 years
9
Key Findings - General
■
Window in which data traffic increases to
contribute 90% of overall traffic is 10 years
from 1997.
■
Crossover point in 1998.
■
Suddenness of the change, as opposed to
change itself, is key in determining post
“crossover” industry structure.
10
Key Findings - IP Telephony
■
Due to short transition interval, growth of IP
telephony will be very sudden and very
significant
■
Potential of infrastructure sharing is key
driver of IP telephony, not bandwidth
savings
11
IP.Phusion Technologies, Inc
■
Mission: To be the premier provider of Open
and Cross-Platform Support Systems for IP
Telephony
–
–
–
“Shrink-wrap” Software: Billing and NMS
Customized Software: Billing/CDR Interfaces
Solutions: Consulting, System Integration, and Network
roll-outs for Service Providers.
Founded: By Researchers from ITC and LCS
■ Contact Info: www.ipphusion.com
■
info@ipphusion.com
12