OpenTwin: An open-source framework for workflow orchestration with enhanced simulation and data management
- Speaker(s): Peter Thoma
- email: peter.thoma@fra-uas.de
Downloads
Abstract
Due to cost constraints, many small and medium-sized enterprises (SMEs) increasingly rely on open-source tools for chip design and related applications. While these tools have reached a high level of maturity and are sufficient for a broad range of use cases, their integration into complex design workflows remains challenging—particularly in collaborative settings where multiple engineers work simultaneously on the same design. A major limitation is the lack of integrated, cross-tool version management. Although general-purpose systems like Git can be employed to a certain extent, they often prove inefficient when handling large volumes of data, such as simulation results. A promising solution lies in the adoption of database-driven, distributed integration platforms that augment existing open-source tools with missing functionalities. In this context, we present OpenTwin (www.opentwin.net), a novel, free and open-source platform originally developed for multiphysics simulation workflows. OpenTwin utilizes MongoDB, a document-based NoSQL database system well-suited for managing large, heterogeneous datasets. At its core, OpenTwin offers a robust version control mechanism applicable to arbitrary data types, enabling efficient management of outputs from external tools. Built on a microservice architecture, the platform supports distributed, team-oriented workflows and provides essential simulation infrastructure, including 3D data processing, meshing, and visualization. The goal of this contribution is to initiate a discussion on how OpenTwin can be adapted and extended to meet the specific requirements of chip design processes.
Software
General information
- Main website: https://opentwin.net