What's new in version 10.7.0

It is now possible to use template models. A template model is a model in which one or more template parameters are defined. Template parameters can be bound to any number of block parameters, provided that the type of the block parameter is the same as that of the template parameter.

The following enhancements come with this new feature:

You can switch a block's parameter to be a template parameter and then assign a new or an existing template parameter. See Editing the parameters of a block for more information.

A new dialog box is available for editing template parameters. See Managing template parameters for more information.

A new instance editor is provided. The instance editor allows you to set up different instances of the same model. The blocks in each instance can then use different values for the template parameters. Each instance can be activated, deactivated, or use different run modes, independently. See Using the Instance Editor for more information.

The model manager uses a new type of card for the template models. See The model manager user interface for more information.

See also Models which explains the two types of models that can now be created: models with template parameters and (as in previous versions) models without template parameters. This topic also explains the relevant roles for the new type of model, which can be the same person or different persons: the model author and the instance maintainer.

Models can now use multiple devices in a model with the concurrency level set to more than 1. See also Model execution for different devices.

The handling of old events is more lenient now. Old events are dropped only if an event for the same timestamp or a more recent timestamp has already been processed by the model. This means that an old event might be processed by some models but dropped by other models. See also Input blocks and event timing.

For slow chains, the following status information is now published within the apama_status parameter:

In addition, a message is now written to the correlator log if the slowest chain is delayed by more than 1 second. See Monitoring periodic status for more information.

Activation messages are now written to the correlator log whenever the activation of a chain is started and completed. See Monitoring the model life-cycle for more information.

The version 1 API of the Analytics Builder Block SDK for input and output blocks is deprecated. See the documentation at https://github.com/SoftwareAG/apama-analytics-builder-block-sdk for information on how to migrate your custom input and output blocks to the version 2 API.

Important:
The new features for multiple devices in a model and more lenient handling of old events are disabled if any input and output blocks using the version 1 API of the Analytics Builder Block SDK are added as extensions.

A new topic has been added which gives new users a brief overview of how to add a simple model, and how to view its output. See First steps: Creating your first model.

The Block Reference which is shown in the model editor is now also available from the documentation. The Overview of all blocks provides links to the descriptions of all the blocks in the block reference.

Analytics Builder now supports branding. This means that the branding settings as described in the Cumulocity IoT User guide at https://cumulocity.com/guides/users-guide/enterprise-edition/#branding now also apply for Analytics Builder.

The appearance of several toolbar buttons in the model editor has changed. Especially the following buttons look different now: the button for changing the devices and the button for showing or hiding the grid.