Information Technology Reference
In-Depth Information
main benefit for data intensive applications that
should support high volumes of transactions.
Other advantage of the on-premise software is
that it allows integration with existing software/
hardware resources. Customization is one more
quality characteristic of in-premise software.
CRM systems usually store, handle and process
sensitive private data of customers that should not
in any case leak to competitors. Therefore safety
is an important non functional requirement. Other
important features involve the back-up file storage,
and online any-time, any where immediate access
to the system. Usability is another important fea-
ture for such an application. A customer should be
able to navigate through different functionalities
and access the information he needs easily and
quickly. Prioritizing non functional requirements
is an indicator that will help managers take a deci-
sion regarding the development and deployment
of a system.
Expected Demand: Seasonal demand. This
type of demand is associated with consum-
ers' interest in particular products only dur-
ing a specific period within the calendar
year. For example, Christmas ornaments
and snow ski equipment are subject to sea-
sonal demand.
Expected demand: Temporary effect.
Expected temporary demand may be
caused due to offers, or low prices, or
clearance period.
Expected Demand: Batch processing.
Batch processing demands involve compu-
tational intensive tasks that demand execu-
tion of a series of programs. Usually such
batch processing procedures may be cost,
time consuming or even unfeasible tasks
when in house resources are considered.
Unexpected Demand: Temporal Effect.
The unexpected demand as mentioned by
Klems et al, (2009) is similar to Expected
temporary effect but the demand behavior
cannot be predicted at all or only in short
time in advance.
C. Estimate User Demand
Estimating the expected demand of the applica-
tion is also very important in order to assess costs.
Expected demand is associated with the number
of users. The number of users affects licensing
costs and hardware costs. Licensing costs are
considered for users that access the applications
and make changes of any kind. On the other hand,
for hardware as the number of users increases, the
hardware must also be improved or performance
becomes unacceptable. Centralized database
models present reasonable costs for 5-10 users,
but present exponential growth of costs as the
number of users increases. Distributed models
are a solution to such problems shifting costs to
PCs. Administration fees are also affected by the
number of users as normally one administrator is
considered every 5-10 users.
While estimating the number of users according
to (Klems et al., 2009), one should keep in mind
four types of demand.
For the CRM system seasonal demand refers
to sales and retails periods that usually present
increased demand volume. In that case the number
of in-house users may increase as the sales are
increased. Temporary effect may refer to clear-
ance period or possible relocation that are seldom
events that may cause extra demands. Expected
demands: Batch processing may involve for the
CRM a period that massive advertisements are
shifted. Finally, unexpected demands for the CRM
may occur when a new product of the company
becomes very popular unexpectedly.
FUTURE WORK
Αs future work we aim to evaluate the proposed
model on real world applications deployment and
compare the three alternatives (cloud, on-premise,
Search WWH ::




Custom Search