Cloud Essentials by Kirk Hausman & Susan L. Cook & Telmo Sampaio

Author:Kirk Hausman & Susan L. Cook & Telmo Sampaio
Language: eng
Format: epub
Publisher: John Wiley & Sons
Published: 2013-04-28T16:00:00+00:00

When designing distributed applications, software architects have to take into account the needs for availability and scalability of the solution as a whole. Single points of failure must be avoided at all costs. For instance, what happens if the one computer holding the inventory data for the store were to break? None of the points of sale would be able to access the data and make a sale.

Several techniques can be used to handle availability. Each point of sale could have a cached copy of the data. In this case, if the server hosting the data is not available, a point of sale can still be used to make a sale as long as the data it contains cached is trusted. All sales would be stored locally until the data server is back online. At this point, data from all points of sale can be copied to the data server and synchronized. Of course, this requires changes to the application design and code. Another solution would be to use failover clustering to provide high availability. Failover clustering would not require changing any of the code for the application but comes at a higher hardware cost. For an example of a highly available multitiered application, see Figure 8.6.

FIGURE 8.6 Representation of the POS application as a highly available distributed application

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.