Actions
are the executables, the fundamental integration steps.
Data transfers are one important kind of action (see below). Other actions
can call and execute an external program or an operating system command,
validate data, execute a SQL statement, script, or standard or customized
piece of script code, send or receive an email, manipulate files on local
disks or remote machines, and repeat previous actions, see Action
types for a full list. Customized plugin actions can also be developed,
see Plugin
action overview for more. Actions can also be combined into complex
workflow sequences known as activities (see below). See Actions
overview for more.
Data
transfers 
are actions that transfer data between
a source and a target data source and are thus a particularly important
action type (see below for more on data sources). Data transfers are created
by specifying the source data source, the target data source, then mapping
between the two, using the graphical Mapping
screen. They may form part of a wider activity that can include complex
manipulations including database updates, transformations using the in-built
calculator and scripts (see below). See Data
transfers overview for more.
A
data source 
is a type of data and a means of accessing
that data. The system supports the exchange of data between any combination
of the following data source types: Databases, which are handled via the
JDBC data source type, file type data sources which include text, CSV,
EDI, XML, IDOC, HTML but which may not have to exist as a file. In
addition Terminal sessions may be the target data source of a data transformation,
and other applications may also be supported, see Data source overview
for links to further discussion and Data source types for a summary of
the data source structures supported.
Resources
include JDBC drivers, JMS server details and JMS
message queues. These can be set up and then used throughout the application,
see Resources overview
for more.
Activities
are complex actions designed as workflows. Actions can
be combined into sequences that allow complex manipulations of data to
be performed. Business rules, notifications, decisions, error handling
as well as rollbacks or compensating actions for transactable items and
components, can all be incorporated into an activity workflow. This allows
the data transformation process to be controlled in very specific ways,
allowing automation even of complex business processes. The built-in graphical
workflow designer lets you build and test an activity interactively and
then save it for execution on a local or remote server. See Activities
overview, Transactions
overview for more.
Projects
are groupings of integration components and resources,
organized for efficiency. Project resources include actions and activities,
data sources, event monitors and global variables. Within a project these
resources can be shared and can be copied to other projects using drag
and drop or copy and paste. Projects can be assigned to authorized
users and are deployed to authorized host servers, see Projects
overview.
Web
services 
are modular, self-describing applications
that are fast becoming a standard means of sending and receiving information
over distributed networks. The system allows the use and publication of
web services which can be incorporated into workflow activities and are
defined as part of a project, see Web
services overview for more.
Users 
can be one of three different categories: Standard
users are assigned to specific projects and hosts. Administrators can
view users, projects and hosts, which ensures managing and monitoring
projects on both local and remote hosts is simple and efficient. Administrators
can also update off-line hosts, remote hosts that store their configuration
data locally and do not normally connect to the primary server. Super
Users are special administrators with full system privileges, see Users
overview.
Hosts
are servers running the application that execute actions
and workflows. The system can be configured with just a single host, used
for both testing and deployment for example within a small business, or
with multiple hosts which may be in different physical locations including
test, backup and live servers. Each host understands how to manage and
route communication between itself and other hosts both outside a firewall
and within the corporate network. In general remote hosts connect to a
primary server for their configuration information and this host can be
directly accessed by the system administrator, see Hosts
overview. Remote administration of a host is possible via an Internet
browser, on a machine which does not have this application installed,
see Remote administration
for more.
Execution 
: Having created the required actions or activities, you can
test them on a local server or execute them directly on a specified host.
Actions and activities can be scheduled for later execution either as
a one-off or on a regular basis, or they may be deployed in such a way
that other systems interact to initiate their execution, see below.
Events 
and Event monitoring 
: An action can
be deployed to a production server for immediate execution, can be scheduled
for execution at a specific time, or you can specify particular events
that will initiate particular actions or workflow activities, allowing
complex processes to occur. For example the following events could feed
incoming data to a workflow activity: arrival of a specific file, locally
or via FTP; the arrival of specific JMS events in a queue; or a simple
socket event. The system can also access remote data sources and can schedule
actions on remote servers using a proxy host facility. To handle these
triggers, event properties are created on the production host server either
directly or by assigning the appropriate action to that host, see Event
overview and Event
Monitor overview for further discussion.