Information Technology Reference
In-Depth Information
cycle leads to scope creep, cost overruns, long working hours, and a
generally unhappy, unmotivated team.
Sometimes there is management pressure to rush through the release
of a product. In such cases, it is advisable to guide management with
regard to any compromises that one will have to make that may affect
the safety of the system. Management can then make a collective decision
on the software release dates. It is better to deal with bad information as
early as possible. Keeping quiet under pressure will only complicate the
problem. In fact, it is quite likely that the cost will multiply if the problem
has to be resolved at a customer site rather than in the development lab.
Techno-political factors rear their heads more often when external
consultants and contractors are involved. Because their involvement in
the project is for the duration of the contract, there could be a tendency
on the part of less than professional contractors to ignore long-term
consequences of failure, as the contractors would have moved on by then.
The project manager should create an environment in which each con-
tributor is committed to success, irrespective of the length of his or her
attachment to the project.
Cascading Failures
As the name implies, cascading failures (Figure 1.5) begin with the failure
of a small component, which in turn triggers the failure of other compo-
nents, leading sometimes to the failure of the entire system. Quite often,
such failures are the result of faulty fail-over implementation strategies.
With fail-over, there is an effort to redistribute the load from a failed node,
which results in an excess load on other components beyond their
capacity. This may cause them to fail and may further aggravate the
situation. The effect snowballs and soon the entire network may also fail.
Failures can cascade
Figure 1.5
Cascading failures.
