We conducted these system tests in January 2026 to evaluate a Reference Architecture for a Land Information Management System with:
Parcel management is often a core component of a Land Information Management System (LIMS); however, the workflows tested here represent only a subset of the broader business processes a full LIMS can support. For this reason, we refer to the system tested as a Parcel Management System, which is a type of Land Information Management System.
We designed the system presented in this test study at a physical level. It incorporates several design choices that are described in detail in the design choices and considerations section. We designed the system to support an organization supporting a volume of typical parcel management workflows representative of a parcel management organization, like a local, state, or federal government, utility, or agricultural company.
We deployed and tested the system for both performance and end-user experience. The load was determined in multiples of operations per hour, or how many times each workflow is executed in an hour. You can read details about this approach in the workflow pacing section.
You can also learn more about test studies at a high level to gain important context.
You should become familiar with the reference architecture before this test study, which has important information and resources that are not duplicated here.
A foundational Parcel Management System delivers the following capabilities, as defined in the data editing and management system pattern, including:
The system capabilities are delivered through the following software with all available patches applied:
The design was deployed on virtual machines in Microsoft Azure with a Microsoft Windows operating system.
The software versions listed above were the latest available when this system test study was deployed and tested. Esri strongly recommends using the latest available software versions when possible.
For these tests, we used a 5 GB parcel fabric dataset with a geographic extent of roughly 2,000 square miles. It contains about 1.5 million tax parcels, 30,000 subdivision boundaries, and 50,000 lots. This dataset is governed by a set of attribute rules that automatically populate parcel-related attributes such as tax year, dimensions, names, and identifiers, as well as enforcing data consistency.