The SUPER Project

The SUPER project is a broadly-based SciDAC institute with expertise in compilers and other system tools, performance engineering, energy management, and resilience. The goal of the project is to ensure that DOE’s computational scientists can successfully exploit the emerging generation of high performance computing (HPC) systems. This goal is being met by providing application scientists with strategies and tools to productively maximize performance, conserve energy, and attain resilience.
Director: Bob Lucas

Recent Accomplishments

Multi-Objective Optimization of HPC Kernels for Performance, Power, and Energy (Balaprakash, Hovland, Wild, and Tiwari)
FUSED: An Application-level Soft Error Detection Framework (Sharma, Gopalakrishnan, and Rakamiric)
Accelerating Block Eigensolvers in MFDn (Aktulga, Williams, and Buluc)
Programming Model Extensions for Resilience at Extreme Scale (Hukerikar, Lucas, and Diniz)
Green Queue: Customized Large-scale Clock Frequency Scaling (Carrington and Tiwari)
Fault Resilience of an Algebraic Multi-Grid (AMG) Solver (Casas Guix and de Supinski)
Hierarchical Auto-Tuning for Lattice Boltzmann Simulation (Oliker and Williams)
SUPER Autotuning with Orio Speeds up PETSc by up to 100% (B. Norris and A. Mametjanov, ANL)
SUPER Tools Diagnose and Quantify Impact of Performance Variability (P. Worley (ORNL) and K. Huck (Univ. of Oregon))
Optimization of Particle-to-Grid Methods on Multicore Systems (Oliker and Williams, also supported by XStack)
Preliminary Performance Tuning for NWChem Texas Integrals (Bert de Jong, Lenny Oliker, Hongzhang Shan)