Research Software Sustainability for the Open-Source Particle Visualization Framework MegaMol

该提案有四个主要目标，以确保可视化框架MegaMol的可持续性。MegaMol是一个以gpu为中心的框架，用于粒子数据的交互式可视化。它是在CRC 716的背景下开发的，但同时也用于其他研究项目。本项目将考虑可视化软件的具体需求及其对可持续性的影响：一致的交互性和良好的可用性都对框架的成功采用做出了重要贡献。这意味着额外的工作（也包括维护），而模拟软件就不是这样。最初，我们专注于为MegaMol的开发、发行和文档完全重做基础架构。我们希望过渡到免费的、基于云的服务和现有基础设施的组合，并从用户和领域科学家那里抽象出来，以简化MegaMol的部署和使用。通过转向公共平台，我们想强调MegaMol的开放性，以及它的发展，以提高其传播。为了改善传播，我们将在能见度高的特定地点或特定基础设施安装MegaMol。通过这种方式，我们希望向更多的开发人员和领域科学家展示使用MegaMol的好处。这项措施将由MegaMol的出版物和开发博客加以补充。为了使它对更广泛的用户群更有用，我们将重新设计MegaMol的用户界面，作为这个项目的第三个方面。新界面将采用集成数据的元分析以及任务特定视图到UI上，以确保用户只以清晰的方式看到相关设置。HTML5将成为这个用户界面的技术基础，以统一交互范例以及MegaMol的运行时行为，而不管使用场景（桌面、集群、远程、powerwall）。最后一个工作包涉及项目本身的公开文档，包括已解决和未解决的挑战。我们想调查需要对斯图加特大学的基础设施进行哪些更改，以及在使用异构资源时出现哪些问题，特别是考虑到冗余和可用性更改。成本文件应当能够对所需资金进行较长期的估计，以确保MegaMol的可持续性。

This proposal has four main goals to ensure the ensure the sustainability of the visualization framework MegaMol. MegaMol is a GPU-centric framework for the interactive visualization of particle data. It is developed in context of the CRC 716, but meanwhile also employed in other research projects. This project will take into account the specific requirements of visualization software and the consequences these have on sustainability: consistent interactivity as well as good usability both contribute significantly to the successful adoption of a framework. This means additional effort (also for maintenance), which is not the case for, e.g., simulation software.Initially we focus on completely reworking the infrastructure for MegaMol development, distribution and documentation. We want to transition to a combination of free, cloud-based services and existing infrastructure and abstract that from the users and domain scientists to simplify deployment and usage of MegaMol. By switching to public platforms we want to emphasize the openness of MegaMol as well as its development to improve its dissemination.To improve dissemination, we will install MegaMol at specific sites with high visibility or particular infrastructure. That way, we want to demonstrate the benefits of employing MegaMol to more developers as well as domain scientists. This measure will be complemented by publications and a development blog for MegaMol.To make it more useful for a broad user base, we will redesign the user interface of MegaMol as a third aspect of this project. The new interface will take meta analysis of ensemble data as well as task-specific views onto the UI to ensure that users only see the relevant settings in a clear way. HTML5 will be the technical foundation of this user interface to unify the interaction paradigm at well as the runtime behavior of MegaMol, regardless of the usage scenario (desktop, cluster, remote, powerwall).The last work package concerns the public documentation of the project itself, including solved and unsolved challenges. We want to investigate which changes are required with regard to the infrastructure of the University of Stuttgart and which problems arise when employing heterogeneous resources, especially considering redundancy and changes to availability. The documentation of costs shall enable a longer-term estimate of required funding to ensure the sustainability of MegaMol.