With FPGAs now being deployed in the cloud and at the edge, there is a need for scalable design methods that can incorporate the heterogeneity present in the hardware and software components of FPGA systems. Moreover, these FPGA systems need to be maintainable and adaptable to changing workloads while improving accessibility for the application developers. However, current FPGA systems fail to achieve modularity and support for multi-tenancy due to dependencies between system components and the lack of standardised abstraction layers. To solve this, we introduce a modular FPGA operating system – FOS, which adopts a modular FPGA development flow to allow each system component to be changed and be agnostic to the heterogeneity of EDA tool versions, hardware and software layers. Further, to dynamically maximise the utilisation transparently from the users, FOS employs resource-elastic scheduling to arbitrate the FPGA resources in both time and spatial domain for any type of accelerators. Our evaluation on different FPGA boards shows that FOS can provide performance improvements in both single-tenant and multi-tenant environments while substantially reducing the development time and, at the same time, improving flexibility.

中文翻译：

---

低聚果糖

随着 FPGA 现在部署在云端和边缘，需要一种可扩展的设计方法，可以整合 FPGA 系统的硬件和软件组件中存在的异构性。此外，这些 FPGA 系统需要可维护并适应不断变化的工作负载，同时提高应用程序开发人员的可访问性。然而，由于系统组件之间的依赖关系以及缺乏标准化的抽象层，当前的 FPGA 系统无法实现模块化和对多租户的支持。为了解决这个问题，我们引入了模块化 FPGA 操作系统——FOS，它采用模块化 FPGA 开发流程，允许更改每个系统组件，并且不受 EDA 工具版本、硬件和软件层的异构性的影响。进一步，为了对用户透明地动态最大化利用率，FOS 采用资源弹性调度来为任何类型的加速器在时域和空间域中仲裁 FPGA 资源。我们对不同 FPGA 板的评估表明，FOS 可以在单租户和多租户环境中提供性能改进，同时显着缩短开发时间，同时提高灵活性。