Carriers
today who are seeking to build new IPTV networks face a number of architectural
questions and issues. A quandary arises due to the fact that technical solutions
will always have specific pros and cons, and their own supporters and detractors;
assuming, of course, that each solution meets the technical requirements of the
service provider. How then can business managers transcend the technical arguments
and make a truly objective network decision that aligns with their business objectives?
This article describes a series of modular bottom-up models which Juniper Networks
has developed with STEM to help carriers to evaluate different technical solutions
for their networks.
The business impact of future uncertainty
One approach to complex technical decision-making is to evaluate implementations
based on business considerations such as total cash flow, profitability, and margin.
Regardless of the number of viable technical approaches, only one will provide the
superior business value. However, assessing business value is dependent on the likelihood
of future, unpredictable events, which is why models are critical to this process.
There are many trade-offs that must be understood prior to making network decisions.
For example, what happens to network costs as a result of:
- centralizing or distributing network intelligence?
- using different placements of network servers and equipment?
- adding or by-passing aggregation equipment?
- implementing multicast either in the router, aggregator or DSLAM?
- increasing take rates of existing and new applications?
- increasing the number of high definition channels viewed?
- increasing VoD concurrency rates?
- increasing TV control traffic in the network?
To fully assess business value, carriers should not only understand how the answers
to these and many other questions affect capital investment but, more importantly,
how they affect total cash flows adjusted for risk. This means understanding which
solution provides the:
- highest operating cash flows
- lowest labour costs (or highest labour productivity and efficiency)
- highest revenue potential
- lowest risk to the business.
Answering these questions seems difficult enough, but the task is even more arduous
given that each question requires numerous variables and inputs to be evaluated.
By building an IPTV model in STEM, Juniper is helping carriers to make a rapid and
reliable evaluation of these questions.
STEM enables comparative bottom-up business analysis of IPTV networks
The Juniper model comprises three modules, preserving a clear separation between
demand and cost elements:
- service and traffic demand module
- a capital investment module for each architecture considered
- operational expense modules.
Part of the traffic demand module, depicting the traffic from the metro or core
network to the Gigabit Ethernet port of the DSLAM (Source: Juniper Networks)
The figure above shows a part of the traffic demand module, which depicts the traffic
from the metro or core network to the Gigabit Ethernet port of the DSLAM. This maps
the traffic flows generated by the services shown on the left to a Broadband Service
Router (BSR) connecting upstream to the metro or core, and downstream either to
the aggregation layer, or directly to the DSLAMs.
The number of circuits generated by the Transformations on the right drive the capital
investment modules for each of the alternative architectures. For example, if the
architecture excludes the aggregation layer, only the DSLAM traffic and upstream
core traffic will feed into it. Alternatively, where this is an aggregation layer
and BSR, all three streams will be sent to the solution.
Separate capital investment and operational expense modules are built for each of
the technical solutions considered. However, all of them are driven by the same
service and traffic demand module. By using a single traffic generator to drive
all the equipment deployments, Juniper ensures a true like-for-like comparison of
the total capital investment needed for each approach, and thus assesses the ability
of the solution to scale over time.
Each capital investment module is based on a set of deployment rules which are specific
to each vendor. Capital investment is driven by:
- chassis bandwidth, subscriber and line card capacities
- ports per line card (or other line card constraints)
- 1:N line card protection rules
- redundancy requirements (chassis, fabrics, router processors, etc.)
- number of sites for each item of equipment
- component pricing.
Once the level of required capital investment is determined, the next and most critical
step is to determine what the carrier will get in return for that investment. To
compare how different implementations influence profit margins, Juniper uses Activity
Based Costing (ABC) to drive the operational expense modules, and then we can work
with the carrier to develop either cost-plus or market-based pricing strategies.
Using Excel-based algorithms developed at Juniper as well as industry benchmarks,
the operational costs for each solution are forecast, and the results input into
the appropriate rental, connection, provisioning, maintenance and administrative
costs for services and equipment resources in STEM. For operational costs, the models
calculate:
- running costs (power, space and equipment maintenance contracts)
- labour or work centre costs (customer care, network management)
- marketing expenses (acquisition, churn, etc.)
- professional and engineering costs (testing, installation, site prep, etc.)
- remediation (security costs).
This enables a carrier to evaluate the operating cash flows that a particular approach
will offer for a specific level of capital investment under a particular set of
assumptions. Carriers can then make business-driven network decisions based on their
comfort level and assessment of the risk of the assumptions, and the financial results
of the analysis for each solution.
Jack Barrett is a Senior Manager at Juniper Networks. He is responsible for developing
and articulating the financial impacts of Juniper’s complete product and solution
portfolio for customers on a global basis. Prior to Juniper, he held positions from
Lead Systems Engineer to Managing Director at AT&T Bell Labs, AT&T, Lucent,
Realtech Consulting, and Unisphere Networks. Jack holds degrees from Rutgers University
(BSEE), University of Southern California (MSEE) and Monmouth University (MBA).
He is currently a member of the New Jersey Chapter of the Institute of Management
Accounts (IMA) and is seeking his certification as a Certified Management Account
(CMA).
Juniper Networks provides purpose-built networking
and security solutions over high-performance IP platforms. Its customers include
major service providers, enterprises, governments, and research and education institutions
worldwide.