Advancements in High-Performance Computing and Benchmarking

The field of High-Performance Computing (HPC) is experiencing significant advancements in benchmarking and performance analysis. Researchers are developing innovative methods to reduce the computational cost of benchmarks while maintaining ranking stability, and new frameworks are being proposed to enable portable and efficient sampling across simulators and real hardware. The integration of cutting-edge technologies, such as exascale supercomputers and novel storage solutions, is also accelerating scientific discovery. Noteworthy papers include:

  • Nugget, a flexible framework for portable sampling that cuts interval-analysis cost by orders of magnitude.
  • BISection Sampling, a novel approach for reducing benchmarks while maintaining stable rankings, which reduces computational cost by up to 99%.
  • Aurora, Argonne's pioneering Exascale supercomputer, which leverages Intel's oneAPI programming environment and integrates the Distributed Asynchronous Object Storage (DAOS) solution.

Sources

Portable Targeted Sampling Framework Using LLVM

Efficiently Ranking Software Variants with Minimal Benchmarks

Aurora: Architecting Argonne's First Exascale Supercomputer for Accelerated Scientific Discovery

An HPC Benchmark Survey and Taxonomy for Characterization

Noise Injection for__Performance Bottleneck Analysis

Built with on top of