|
THE Distributed Capacity Management Section
(DCMS) is the group within the IRS responsible
for this area of activity. According to
DCMS, the greatest benefit of Athene is
its modeling capability, together with
the data collection and reporting capabilities.
Before Athene was implemented, a majority
of staff time was spent collecting data,
analyzing it and producing manual reports.
Now, only about 10% of their time is spent
in this way, and the rest is devoted to
growth strategy and trend analysis. A side
benefit is that no more tedious spreadsheets
have to be produced by hand. DCMS take graphs
produced by Athene and use them for reports
to their upper management, as well as reports
to operations. They also take Athene reports
and publish them, with graphs, on the IRS
internal web site to increase understanding
within the Agency.
Athene was used to assess whether the IRS
needed to buy a faster system with more
capacity to handle their Automated Under
Reporter system or whether adjustments
could be made to the existing system. modeling
of an increased workload and how it could
be handled was carried out, using Athene.
Marty Leonka, Section Chief: "Athene allows
you to model what will happen without spending
a lot of time, resources and money."
|
With the introduction
of Electronic Federal Tax Payment (providing
electronic filing for businesses), DCMS
were keen to establish whether it was necessary
to invest in a new six-engine mainframe
computer which was being promulgated as
the solution to increased processing requirements.
The situation was modeled using Athene:
results indicated that with the use of multi-threading
(as opposed to the existing single threaded
approach), transaction processing could
be achieved at the requisite speed with
just a few minor software improvements.
Athene is used to provide monitoring of
a Telephone Routing System which has to
be operational 24/7. A fault was identified
by Athene, and remedial work undertaken
immediately before users were aware of the
problem.
Athene identified
that within one group of workers commands
were being sent once per second, compared
with the average of once every 15 minutes.
The individual user was located - and it was found that a stack of
books were leaning against the keyboard -
the source of the 'every second' commands!
Next
Case Study |
