IS1060 Introduction to information
systems
64
then
access it as they need to, probably by using networks. The alternative
is to distribute the database. We could store, for example, all data relating
to motor
cars in England in one place, and all data relating to Scotland
in another. Logically it might be one database, but operationally data is
stored closer to where it is used. Software might take care of all this detail
as seen by any user – in the case of photos
stored on Picasa or Dropbox
you and I have no idea where in the world they really are, or where there
are backup copies of the photos! We just trust the service supplier.
We may also choose to replicate a database, with a full copy of the
database held in two or more locations. This
could be a way to ensure
security and integrity. If one datacentre is out of action, another is
available, but there is the problem of ensuring that updates and changes
made in one copy are reproduced on the others and that the copies remain
synchronised. Again, software can help, but this is more complex to do and
may at times fail.
4.7 Reminder of learning outcomes
Having completed this chapter, and the Essential reading and activities,
you should be able to:
• express a logical understanding of how the technical parts of a
computer-based information system work,
their principal structures
and components including contemporary software technologies for
information processing and communications
• demonstrate a good understanding of the significance of history for
understanding contemporary information systems and the concept of
legacy systems
• discuss the evolution of different types of information and
communication technologies (eras) and the extent to which new
technologies have led to changes in the way organisations use
technology and are structured and operate
• explain client–server, enterprise and
cloud computing and give
examples of each
• describe the database approach and offer examples of its advantages
over a file-based approach.
Dostları ilə paylaş: