Transcript ebiTrack Architecture

ebiTrack Architecture
Version 1.0
September 24, 2012
Overview
• The ebiTrack implementation consists of the
following components: a native iPad
application; a RESTful web service; a standalone SQLServer database; database triggers
that are added to tables in the BackTrack
database(s).
• The server-side components (database and
web service) run on a Windows™ server and
requires SQL Server and IIS to be installed.
Overview Continued
• ebTrack co-exists with an existing BackTrack
installation and requires no changes to the
BackTrack system or database. The only
addition is triggers on certain BackTrack tables
to allow for reducing the overhead of refresh
operations on the iPad client. (Basically we
track the last time a change was made to a
table so the iPad only needs to update the
relevant table in its local storage).
Web Service
• A single web service, with multiple public
methods, provides the interaction between the
iPad client and the BackTrack database. This is a
.Net RESTful web service that runs under the
context of IIS. Site-specific options are defined in
the web configuration file. All database access is
done via stored procedures and no database
credentials are passed from the iPad to the web
service so there is no risk of exposing credentials
should an iPad ‘disappear’. Each registered iPad
has a separate context maintained and updated
each time a transaction is processed.
ebiTrack Database
• The web service uses a ebi developed database to
maintain context and credentials for the iPad
application. This database also contains all the
stored procedures that are used to read and write
data from the existing BackTrack databases.
Support for all the ebi feature additions is
provided from this database and not from the
BackTrack databases to insure forward and
backward compatibility. The web service only
executes procedures from within the ebiTrack
database.