Why the Next Generation of HPC systems Needs Open Source Driven Standardization
summary
Panel argues that ARM64 is ready for primetime with ACPI/UEFI enablement, off‑the‑shelf systems and cloud instances (AWS A1) now available. Work focuses on concrete items: GCC/LLVM and library optimizations, SVE vector support and whether SVE uses fixed or variable length, and open SVE toolchain and runtime availability. A shared lab virtualizes hardware with KVM, SR‑IOV and emulation to run OpenHPC functional and upcoming performance regressions. Packet highlights multi‑architecture bare‑metal delivery as a production channel. SVE needs open toolchain support
AI-written summary · about
chapters (8)
- 0:00Introduction and panel setup
- 1:53ARM in HPC and open source role
- 5:47Software ecosystem enablement
- 11:53RIKEN’s Post-K and SVE optimizations
- 17:53Packet’s heterogeneous cloud approach
- 21:04Shift in HPC compute delivery
- 23:19AI enablement in supercomputing
- 26:11Closing remarks and thanks
related talks
01h 01mPanel: Building a Scalable RISC-V Software Ecosystem
13mKeynote: Open Source: The New Normal
12mOpenStack for Computational workloads
47mPrepping your Open Source Project for Multi-Platform
24mEuropean Processor Initiative
01h 13mWelcome & Keynote Sessions
17mOpenISA Miniconf Introduction
41mRISC-V and Open Source Hardware BoF
