Using Dictionaries to Manage Data Within a Modeling Framework System

Introduction

Under contracts with a number of federal agencies, the Pacific Northwest National Laboratory developed a modeling FRAMEwork System (FRAMES) to support the linkage, integration, and communication between multiple models and databases (Whelan et al.  1997,  PNNL-11748).Consisting of a suite of integration tools and a user-friendly modeling platform, FRAMES ensures that data are correctly transferred among databases and models, especially those components developed with different compilers, programming languages, and database packages.FRAMES was designed to allow models and databases to be accessed from both PCs and mainframes, maintaining the flexibility of access as the components are updated over time.

FRAMES is a flexible and versatile system intended to be relatively "open".  That is, it is relatively inclusive of models, attributes, databases, and other software.This inclusivity requires that data be managed effectively.FRAMES provides a data management approach that uses DICtionaries (DICs), which are collections of information describing the data to be managed.Examples of such information include parameter name, description, units, and ranges.This approach of using DICs frees developers to focus on how models and databases consume and produce data, while providing an easily understandable process for those who use the system to simulate real-world problems.

The following sections provide background on the development of this unique approach, describe the purpose and functions of dictionaries used by the system, and provide information and examples for model or database developers interested in creating dictionaries for use in FRAMES.The appendix provides a full set of DICs that comes with FRAMES.