Class Selection Strategies

Due to the complex nature of AVEVA™ E3D Design and the associated project data, it is advised to create the configurations so as to ensure that the publishing performance of the AVEVA™ Gateway for Unified Engineering remains optimal.

Models in AVEVA™ E3D Design could often have large number of sub classes under a parent class (for example, SITE). For such projects, publishing the engineering and graphical data of an entire SITE as a single model (single XGL/ZGL file) can result in memory-related issues and sub optimal performance of the AVEVA™ Gateway for Unified Engineering. Therefore, it is advised to create configurations in such ways that generated models would be split into multiple files of smaller sizes, for example by ZONE elements or by disciplines.

Consider a configuration that has the SITE class selected as the topmost element for both engineering and graphical data as shown in the following image.

If you have selected a SITE (for example, SITE-PIPE-AREA01) for publishing with the above configuration, the AVEVA™ Gateway for Unified Engineering would generate one EIWM file with the engineering data and one XGL/ZGL file with the graphical data. Such configuration may run without any memory issues if the selected SITEs are not too large and complex in graphical model. However, very often a SITE has dozens or even hundreds of ZONEs which are heavy in 3D model on their own.

The alternative approach is to select the lower classes under SITE as the sources of graphical data as shown in the following image.

Selecting a SITE (for example, SITE-PIPE-AREA01) for publishing with such configuration would still generate one large EIWM file for the engineering data but the graphical data would be separated into multiple files based on number of ZONEs.

If you have selected a ZONE (for example, ZONE-PIPE-AREA01) for publishing with the same configuration, then gateway would generate the same number of EIWM and XGL/ZGL files based on number of ZONEs. Such strategy of selection would have much less impact on memory usage during the publishing process.

Dividing models into smaller parts for publishing is also beneficial in case of future model updates. Consider the case where a large SITE has been published as one model to AVEVA™ Asset Information Management and at a later point in time, it must be updated due to changes in source model. In such cases, the entire SITE need not be published once again if the graphical data has been split into multiple files.