Results

Submissions: 69 (70% of accepted papers) including 2 major revision rejections

Evaluation Results:

• 66 Artifact Available (+2 rejected major revision)
• 52 Artifact Functional (+2 rejected major revision)
• 38 Results Reproduced (+2 rejected major revision)

SC21 Best Reproducibility Advancement Award Finalists:

• Weihao Cui, Han Zhao, Quan Chen, Ningxin Zheng, Jingwen Leng, Jieru Zhao, Zhuo Song, Tao Ma, Yong Yang, Chao Li, Minyi Guo: “Enable Simultaneous DNN Services Based on Deterministic Operator Overlap and Precise Latency Prediction”
• Ankit Srivastava, Sriram Chockalingam, Maneesha Aluru, Srinivas Aluru: “Parallel Construction of Module Networks”
• William S. Moses, Valentin Churavy, Ludger Paehler, Jan Hückelheim, Sri Hari Krishna Narayanan, Michel Schanen, Johannes Doerfert: “Reverse-Mode Automatic Differentiation and Optimization of GPU Kernels via Enzyme”
• Sunil Kumar, Akshat Gupta, Vivek Kumar, Sridutt Bhalachandra: “Cuttlefish: Library for Achieving Energy Efficiency in Multicore Parallel Programs”
• Konstantinos Parasyris, Giorgis Georgakoudis, Harshitha Menon, James Diffenderfer, Ignacio Laguna, Daniel Osei-Kuffuor, Markus Schordan: “HPAC: Evaluating Approximate Computing Techniques on HPC OpenMP Applications”

Paper title	Available	Functional	Reproduced	Artifact
Temporal Vectorization for Stencils				Artifact
LCCG: A Locality-Centric Hardware Accelerator for High Throughput of Concurrent Graph Processing				Artifact
Parallel Construction of Module Networks				Artifact
Online Evolutionary Batch Size Orchestration for Scheduling Deep Learning Workloads in GPU Clusters				Artifact
Krill: A Compiler and Runtime System for Concurrent Graph Processing				Artifact
Exploiting User Activeness for Data Retention in HPC Systems				Artifact
LogECMem: Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging				Artifact
KAISA: An Adaptive Second-order Optimizer Framework for Deep Neural Networks				Artifact
Understanding, Predicting and Scheduling Serverless Workloads under Partial Interference				Artifact
Cuttlefish: Library for Achieving Energy Efficiency in Multicore Parallel Programs				Artifact
Accelerating XOR-based Erasure Coding using Program Optimization Techniques				Artifact
Accelerating Applications using Edge Tensor Processing Units				Artifact
On the Parallel I/O Optimality of Linear Algebra Kernels: Near-Optimal Matrix Factorizations				Artifact
SV-Sim: Scalable PGAS-based State Vector Simulation of Quantum Circuits				Artifact
DeltaFS: A Scalable No-Ground-Truth Filesystem For Massively-Parallel Computing				Artifact
Reducing Redundancy in Data Organization and Arithmetic Calculation for Stencil Computations				Artifact
CAKE: Matrix Multiplication Using Constant-Bandwidth Blocks				Artifact
Hardware-supported Remote Persistence for Distributed Persistent Memory				Artifact
STM-Multifrontal QR: Streaming Task Mapping Multifrontal QR Factorization Empowered by GCN				Artifact
Clairvoyant Prefetching for Distributed Machine Learning I/O				Artifact
Discovering and Balancing Fundamental Cycles in Large Signed Graphs				Artifact
High Performance Uncertainty Quantification with Parallelized Multilevel Markov Chain Monte Carlo				Artifact
3D Acoustic-Elastic Coupling with Gravity: The Dynamics of the 2018 Palu, Sulawesi Earthquake and Tsunami				Artifact
Enable Simultaneous DNN Services Based on Deterministic Operator Overlap and Precise Latency Prediction				Artifact
Index Launches: Scalable, Flexible Representation of Parallel Task Groups				Artifact
PAGANI: A Parallel Adaptive GPU Algorithm for Numerical Integration full strip note				Artifact
Minimizing privilege for building HPC containers				Artifact
Efficient Tensor Core-based GPU Kernels for Structured Sparsity under Reduced Precision				Artifact
Preparing an Incompressible-Flow Fluid Dynamics Code for Exascale-Class Wind Energy Simulations				Artifact
Efficient Large-Scale Language Model Training on GPU Clusters				Artifact
Pinpointing Crash-Consistency Bugs in the HPC I/O Stack: A Cross-Layer Approach				Artifact
AgEBO-Tabular: Joint Neural Architecture and Hyperparameter Search with Autotuned Data-Parallel Training for Tabular Data full strip note				Artifact
Bootstrapping In-situ Workflow Auto-Tuning via Combining Performance Models of Component Applications full strip note				Artifact
Systematically Inferring I/O Performance Variability by Examining Repetitive Job Behavior				Artifact
MAPA: Multi-Accelerator Pattern Allocation Policy for Multi-Tenant GPU Servers				Artifact
Productivity, Portability, Performance: Data-Centric Python				Artifact
Flare: Flexible In-Network Allreduce				Artifact
Non-Recurring Engineering (NRE) Best Practices: A Case Study with the NERSC/NVIDIA OpenMP Contract				Artifact
SEEC: Stochastic Escape Express Channel				Artifact
APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores				Artifact
Paths to OpenMP in the Kernel				Artifact
The Hidden cost of the Edge: A Performance Comparison ofEdge and Cloud Latencies				Artifact
Scalable adaptive PDE solvers in arbitrary domains				Artifact
ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning				Artifact
Accelerating large scale de novo metagenome assembly using GPUs				Artifact
Online Optimization of File Transfers in High-Speed Networks				Artifact
In-Depth Analyses of Unified Virtual Memory System for GPU Accelerated Computing				Artifact
HatRPC: Hint-Accelerated Thrift RPC over RDMA				Artifact
Characterization and Prediction of Deep Learning Workloads in Large-Scale GPU Datacenters				Artifact
Reverse-Mode Automatic Differentiation and Optimization of GPU Kernels via Enzyme full strip note				Artifact
LibShalom: Optimizing Small and Irregular-shaped Matrix Multiplications on ARMv8 Multi-Cores				Artifact
Simurgh: A Fully Decentralized and Secure NVMM User Space File System				Artifact
HPAC: Evaluating Approximate Computing Techniques on HPC OpenMP Applications.				Artifact
Ribbon: Cost-Effective and QoS-Aware Deep Learning Model Inference using a Diverse Pool of Cloud Computing Instances				Artifact
Lunule: An Agile and Judicious Metadata Load Balancer for CephFS				Artifact
A Next-Generation Discontinuous Galerkin Fluid Dynamics Solver with Application to High-Resolution Lung Airflow Simulations				Artifact
LMFF: Efficient and Scalable Layered Materials Force Field on Heterogeneous Many-Core Processors				Artifact
Whale: Efficient One-to-Many Data Partitioning in RDMA-assisted Distributed Stream Processing Systems				Artifact
ndzip-gpu: Efficient Lossless Compression of Scientific Floating-Point Data on GPUs				Artifact
cuTS: Scaling Subgraph Isomorphism on Distributed Multi-GPU Systems Using Trie Based Data Structure				Artifact
Representation of Women in High-Performance Computing Conferences				Artifact
PEPPA-X: finding program test inputs to bound silent data corruption vulnerability in HPC applications				Artifact
Tensor processing primitives: a programming abstraction for efficiency and portability in deep learning workloads				Artifact
DistGNN: scalable distributed training for large-scale graph neural networks				Artifact
Dr. Top-k: delegate-centric Top-k on GPUs				Artifact
Hybrid, scalable, trace-driven performance modeling of GPGPUs				Artifact
Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines