By using Systems Group RunTime, you can easily:
- Validate the fundamental infrastructure functionality of ETHZ-HACC, ensuring its reliability and performance.
- Expedite project creation through intuitive multiple-choice dialogs and templates, streamlining the development of your accelerated applications.
- Seamlessly integrate with GitHub CLI for efficient version control and collaboration.
- Effectively manage ACAPs, FPGAs, multi-core CPUs, and GPUs, all through a unified device index.
- Transition between Vivado and Vitis workflows effortlessly, eliminating the need for system reboots and enhancing your development agility.
- Design, build, and deploy modern FPGA-accelerated applications with ease, leveraging your own or third-party integrations, for instance, ACCL, Coyote, or EasyNet (please refer to our Infrastructure for heterogeneous architectures and hardware acceleration).
- Explore model-based design principles with readily available out-of-the-box examples.
- Simplify the creation of HIP/ROCm GPU applications using the sgutil new-build-run hip commands.
We use Infrastructure as Code (IaC) and Ansible Automation Platform (AAP) to operate the cluster efficiently. Under the scope of a DevOps methodology, we achieve the following:
- Continuos delivery,
- Avoid manual configuration to enforce consistency,
- Use declarative definition files helping to document the cluster,
- Easily follow Xilinx’s tools versioning release schedule, and
- Deliver stable tests environments rapidly at scale.