Optimizing Virtualized FactoryTalk Architecture for Reliable Power Monitoring

Hargrove Controls & Automation resolves virtual environment challenges to support a power generation facility in stabilizing its FactoryTalk environment and preparing for scalable, distributed access.

OBJECTIVE: The project focused on upgrading the existing FactoryTalk system while transitioning from a network station to a network-distributed architecture. This would allow multiple clients to access HMIs and system data across a distributed environment and support future expansion. The transition also included upgrading to a newer version of FactoryTalk View and modernizing the control environment.

From the facility’s perspective, a successful project would include a smooth transition, improved access from both field and office environments, and the ability to scale without adding additional physical infrastructure.

CHALLENGE: As the system was being updated to support a distributed architecture, issues emerged within the virtual environment. FactoryTalk components, such as the HMI application server and data servers, were operating on the same virtual server, which not only causes resource conflicts, communication challenges, and inconsistent system behavior but also goes against Rockwell standards and best practices. In some cases, components competed for system resources, resulting in unstable communication between services. This configuration did not align with Rockwell and ISA best practices, which recommend separating these components across multiple servers to avoid resource contention and single points of failure.

During the system commissioning, the team encountered uncommon errors when accessing the system through the HMI, including communication and permission-related failures. While the underlying production process remained unaffected, these issues disrupted visibility into the system. These errors had not been documented or encountered by Rockwell’s engineering team, which resulted in the team having to use unique methods to find a solution. After extensive research across multiple PLC tech support groups, it was determined that these errors were, in fact, not an issue caused by the Rockwell configuration or software.

The transition from a network station to a distributed system introduced additional complexity, especially in how different components communicated across the virtual environment. Differences in access methods (specifically between remote desktop access and direct console access) introduced inconsistent behavior during configuration.

BACKGROUND: A Louisiana power generation facility relied on a Rockwell Automation FactoryTalk system for power monitoring and visualization. The existing setup functioned as a network station, where a single server supported a limited number of users and centralized all FactoryTalk components. The system also combined multiple FactoryTalk components — such as HMI Server, Data Server, and Historian — on a single server, which limited flexibility and introduced risk as the system changed over time.

As operational needs evolved, the facility required broader access to system data across multiple users and locations. The team identified an opportunity to reduce the hardware footprint by moving from a traditional physical server setup to a virtualized environment using VMware vSphere.