' Hearing on the Year 2000 Computer Problem and Telecommunication Systems'
'Statement of A. Gerard Roth, Vice President
Technology Programs, GTE
Testimony Before the Subcommittee on Oversight
of the House Committee on Ways and Means
Hearing on the Year 2000 Computer Problem and Telecommunication Systems
June 16, 1998
Chairwoman Johnson and members of the Subcommittee. Good afternoon, my name is A. Gerard Roth. I am Vice
President, Technology Programs, GTE, responsible for GTE's Corporate Year 2000 Program Management Office
(PMO). I am here to discuss Year 2000 (Y2K) as it relates to telecommunications interdependencies and
interoperability. I commend the Subcommittee on Oversight for conducting hearings focused on the telecommunications
industry interdependency and ask that a copy of my written remarks be entered into the record.
Introduction
The focus of this testimony is to bring perspective from GTE's considerable Y2K experience and apply it to this
important discussion of telecommunications interdependency. GTE, and other telecommunications providers, are
continually being asked, "Why can't you simply certify your network and tell us when it will be compliant?" The
following discussion will describe the complexity and interdependencies that make that question so difficult to answer.
Three principal issues come into play in addressing this question:
1. Ownership - There is no one owner of the whole network.
2. Mathematics - The permutations and combinations of calling events, service requests and routing possibilities
exceed the industry's ability to do 100% testing of Y2K.
3. Testing cannot be done on the live network - Out of cycle - clock roll-ahead testing would disrupt current
operations, create unacceptable outages.
Nevertheless, the work being done to remediate and test the Public Switched Telephone Network (PSTN) is well
thought through, is being accomplished at an acceptable rate, and represents the best known solution to Y2K we are
able to accomplish. I believe you will be reassured today that the telecommunications industry understands the
importance of the Y2K challenge and is working aggressively to meet it.
Y2K Background
Year 2000 is unique in the history of this the telecommunications and information technology industries. Left unattended,
it could simultaneously undermine the operation and reliability of the computer and network infrastructure at a specific,
known, future point in time. However, it is also true that the precise impact on our information-based economy and
society cannot be predicted in part due to the complexity and interdependency of systems. Nor can we accurately
predict the full extent of successful remediation of Y2K due to the interoperable, multi-path nature of the PSTN.
We can, however, significantly reduce the likelihood of the apocalyptic scenarios sometimes predicted for January 1,
2000. You will hear from me and my collegues at this hearing some of the measures underway to address your Year
2000 concerns.
Let me quickly summarize some of the major lessons of Y2K that contribute to the complexity:
Schedule is not just important; it is the only thing.
Y2K is truly a "weakest link" problem -- the single system or date conversion we miss may be the undoing of the
99% we did find.
Normally, development and maintenance activities introduce incremental change into an otherwise stable
environment; however, in Y2K, modified systems are reintroduced into an environment which has been
universally and simultaneously de-stabilized.
Since it is impossible to recreate an "off-line" PSTN for testing, complex Year 2000 interoperability must be
tested in pieces by various companies separately and can be actually proven compliant only once those pieces
are in operation together on January 1, 2000.
There is an increasing recognition of the need to devote more time and effort to enterprise and interoperability
testing than was previously planned.
Testing of all types constitutes greater consumes more than 50% of required cost and effort; actual conversion of
applications or products is relatively minor.
Completion of conversion prior to the end of 1998 should be a priority. This allows for validation of year-end
close transactions in the operational environment, and provides up to 12 months of Y2K verification testing.
Test everything you can.
Complexity and Interdependency
In discussing the complexity and interdependency of the PSTN, I intend to gradually "build the onion" from the center
using four models. To demonstrate an increasing complexity and interdependency, each model builds upon the one
before like the currently popular "nesting dolls". The end result points to the conclusion that the PSTN is not readily
certifiable due not only to ownership issues but also to mathematical complexity and test scenario limitations.
1. The Basic System
Telecommunications complexity begins with the essential, computer-based systems used in telecommunications. Chart 1
depicts a representative, physical system, best described as a combination of hardware, firmware, software products
and applications. A typical company will have hundreds of these. Each component in Chart 1 must be assessed for
Y2K impact; remediated, if required, and tested and verified compliant at the system level.
2. The Functional Thread
To perform a "function" (such as customer contact, service provisioning, call routing), these physical systems must work
with others. Chart 2 depicts a real customer service provisioning cluster as an example of system interdependencies.
This example portrays the relationships of among 17 separate systems or users, 26 formal communications channels,
and more than 10 separate data bases needed to provide a basic service. This also includes the actual linkage of
support functions to the physical PSTN and to other companies.
Each of the elements depicted here can also be represented in some version of Chart 1. As such, assuring that each of
the core components is Y2K compliant precedes the verification of this functional thread. A medium-sized company will
have hundreds of these threads.
3. The Core Interoperability of the PSTN - Logical Topology
Chart 3 expands the interdependency model to include network elements of the PSTN.
Logically, each component on this chart can be described functionally as a thread or cluster. This schematic simply
integrates the elements of the legacy software systems. The systems manage the Network, the signaling, data and voice
components of the switched network. This complex hierarchy of systems and interaction of function provides an
example of a single company's network interdependency. The actual LEC portion of the Public Switched Telephone
Network provides for random, multi-path, real-time interaction of these elements, simultaneously processing thousands
of calls each minute of each day.
By way of example, a typical local exchange carrier may have several million digitally switched access lines in the U.S.
Depending upon the LEC, it may have also hundreds of unique systems worldwide, representing millions lines of
computer code (LOC), all of which must be Y2K tested in thousands of functional test clusters. In addition, to verify
Y2K readiness, the typical LEC must assess and test perhaps a couple of thousand of central office (e.g., end office)
and, possibly, international gateway switches and associated support systems in not only domestically but, depending
upon the company, in overseas locations as well.
4. The Expanded PSTN
The final illustration, (the outside of the "onion") depicted in Chart 4, captures the logic of the Public Switched
Telephone Network overall by incorporating the essential logic of Chart 3 for each of several Local Exchange Carriers
(LECs), Inter-exchange Carriers (IXCs), International PTT Interfaces, Customer Premise Equipment (CPE), and
Private Network installations. Within the U.S. alone there are hundreds of local exchange carriers, several
inter-exchange carriers; and perhaps millions of private networks or customer premise emplacements. The obvious
interconnectivity potential of these network relationships points to the mathematical limitations on 100% test. It should
not go unnoticed that the largest external risk to the operational integrity of the PSTN is the continued availability of
electric power across the national power grid on January 1, 2000.
The Call/Virtual Network Circuits
The Nationwide PSTN processes millions of calls per minute. In order to complete a telecommunications transaction
beyond a local exchange, "calls" are spontaneously routed in advance to take advantage of the most efficient
call-processing path. Furthermore, in addition to routine voice/data traffic and wireless access, the PSTN provides a
myriad of additional services including call waiting, directory assistance, 800-number look-up, and 911 emergency
support. Because of this complexity, a "virtual" network circuit is dynamically defined for each transaction such that
advance prediction of specific circuit connectivity is impossible to determine. The process creates a continuously
changing pattern within and between LECs and IXCs. Consequently, it is impossible for any one company to verify the
whole network Y2K compliant.
Nonetheless, the worldwide telecommunications Year 2000 remediation effort is proceeding, with LEC's, IXC's and
others actively and cooperatively working to implement Y2K readiness at all levels of this model. You will hear more
today from my colleagues on this panel on what the industry is doing to respond to Y2K.
Cost
It is useful in closing to give this technical issue a context or grounding with respect to cost. GTE currently expects to
spend about $350 million on Y2K compliance; - more than 50% of which is focused on the testing of these products,
applications, or and the interoperability test of the functions they provide.
As of March 1998, a sample of SEC filings indicates seven (7) telecommunications companies (LECs and IXCs) have
estimated a combined expenditure in excess of $2 billion for Year 2000 remediation. This is a gigantic task; one that I
fear is often trivialized by the casual media and unfairly criticized by otherwise well-intentioned Y2K experts. Nothing is
to be gained by public chanting of doomsday scenarios formed out of ignorance.
It should be noted that this massive Year 2000 remediation activity is occurring at the same time as we continue
operating the live PSTN and maintaining it with the quality of service and variety of features we have all come to take
for granted.
Chairwoman Johnson, I thank you for the opportunity to present this testimony.
house.gov |
