Sequoia logo

ASC Sequoia Benchmark Codes


ASC Sequoia Benchmarks
  Language Parallelism Code Description
Tier Code Fortran Python    C     C++   MPI  OpenMP Pthreads
1 UMT X X X X X X   Marquee performance code. Single physics package code. Unstructured-Mesh deterministic radiation Transport.
1 AMG     X   X X   Marquee performance code. Algebraic Mult-Grid linear system solver for unstructured mesh physics packages.
1 IRS     X   X X   Marquee performance code. Single physics package code. Implicit Radiation Solver for diffusion equation on a block structured mesh.
1 SPhot X       X X   Marquee performance code. Single physics package code. Monte Carlo Scalar PHOTon transport code.
1 LAMMPS       X X     Marquee performance code. Full-system science code. Classical molecular dynamics simulation code (as used).
 
2 Pynamic   X X   X     Subsystem functionality and performance test. Dummy application that closely models the footprint of an important Python-based  multi-physics ASC code.
2 CLOMP     X     X   Subsystem functionality and performance test. Measures OpenMP overheads and other performance impacts due to threading.
2 CLOMP_TM     X     X   Subsystem functionality and performance test. Measures transactional memory, speculative execution, and OpenMP atomic and critical overheads.
2 IOR     X     X   Interleaved or Random I/O Benchmark. IOR is used for testing the performance of parallel file systems using various interfaces and access patterns.
2 Phloem MPI Benchmarks     X   X     Subsystem functionality and performance tests. Collection of independent MPI benchmarks to measure the health and stability of various aspects of MPI performance, including interconnect messaging rate, latency, aggregate bandwidth, and collective latencies under heavy network loads.
2 FTQ/FWQ     X       X Fixed time quantum and fixed work quantum tests. Measures operating system noise.
2 Memory Subsystem Benchmarks X   X     X   Provides the STREAM and STRIDE memory benchmarks to measure various aspects of system memory performance and scalability.
 
3 UMTmk X             Threading compiler test and single CPU performance.
3 AMGmk     X     X   Sparse matrix-vector operations single CPU performance and OpenMP performance.
3 IRSmk     X         Single CPU optimization and SIMD compiler challenge.
3 SPhotmk X             Single CPU integer arithmetic and branching performance.
3 Crystalmk     X         Single CPU optimization and SIMD compiler challenge.

Sequoia Benchmark Glossary

Sequoia Benchmark Results


UMT (Tier 1)

AMG (Tier 1)

IRS (Tier 1)

SPhot (Tier 1)

LAMMPS (Tier 1)

Pynamic (Tier 2)

CLOMP (Tier 2)

CLOMP_TM (Tier 2)

IOR (Tier 2)

Phloem MPI Benchmarks (Tier 2)

FTQ/FWQ (Tier 2)

Memory Subsystem Benchmarks (Tier 2)

UMTmk (Tier 3)

AMGmk (Tier 3)

IRSmk (Tier 3)

SPhotmk (Tier 3)

Crystalmk (Tier 3)


   ASC logo