Co-optimization of flexible Iceberg error-detection gadgets with QAOA via tree search improves success probability and post-selection on Quantinuum H2-1 hardware up to 34 algorithmic qubits.
hub Canonical reference
InProceedings of the International Conference for High Performance Computing, Networking, Storage, and Analysis(Atlanta, GA, USA)(SC ’24)
Canonical reference. 92% of citing Pith papers cite this work as background.
28
Pith papers citing it
Background
92% of classified citations
abstract
We propose a Hilfer advection-diffusion equation of order $0<\alpha<1$ and type $0\leq\beta\leq1$, and find the power series solution by using variational iteration method. Power series solutions are expressed in a form that is easy to implement numerically and in some particular cases, solutions are expressed in terms of Mittag-Leffler function. Absolute convergence of power series solutions is proved and the sensitivity of the solutions is discussed with respect to changes in the values of different parameters. For power law initial conditions it is shown that the Hilfer advection-diffusion PDE gives the same solutions as the Caputo and Riemann-Liouville advection-diffusion PDE. To leading order, the fractional solution compared to the non-fractional solution increases rapidly with $\alpha$ for $\alpha > 0.7$ at a given time $t$; but for $\alpha<0.7$ this factor is weakly sensitive to $\alpha$. We also show that the truncation errors, arising when using the partial sum as approximate solutions, decay exponentially fast with the number of terms $n$ used. We find that for $\alpha< 0.7$ the number of terms needed is weakly sensitive to the accuracy level and to the fractional order, $n\approx 20$; but for $\alpha>0.7$ the required number of terms increases rapidly with the accuracy level and also with the fractional order $\alpha$.
hub tools
citation-role summary
background 12
baseline 1
citation-polarity summary
representative citing papers
NCCLZ decouples quantization and entropy coding across NCCL stack layers to enable overlapped compression, delivering up to 9.65x speedup over plain NCCL on scientific and training workloads.
AtomWorld enables the first direct atomistic simulation of RPV steel at year-and-meter scales, handling ten-quintillion-atom systems and simulating one service year in 1.71 days with 92-97% scaling efficiency on leadership supercomputers.
Mosaic uses cross-modal clusters as the unit for KVCache organization in VLMs to achieve up to 1.38x speedup in streaming long-video understanding.
A planner-executor multi-agent system using gpt-oss-120b and Parsl orchestrates scalable high-throughput MOF screening on the Aurora supercomputer with low overhead.
NestPipe achieves up to 3.06x speedup and 94.07% scaling efficiency on 1,536 workers via dual-buffer inter-batch and frozen-window intra-batch pipelining that overlaps communication with computation.
Simulation study shows cold TLB misses in reverse address translation dominate latency for small collectives in multi-GPU pods, causing up to 1.4x degradation, while larger ones see diminishing returns.
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
Asteria is a runtime system that enables second-order optimization for LLMs by dynamically distributing optimizer state across GPU, CPU, and NVMe while using asynchronous inverse-root computations and bounded-staleness synchronization.
A three-layer co-design optimization using digital twins and surrogate modeling for CDU partitioning and flow control in HPC cooling plants achieves 35.48% annual energy savings, nearly matching the current Frontier design while reducing assignment sensitivity by 93%.
CCCL delivers 1.34-1.94x faster cross-node GPU collectives via CXL memory pooling than 200 Gbps InfiniBand RDMA, with 1.11x LLM training speedup and 2.75x hardware cost reduction.
EditFlow reconstructs temporal developer editing flows from code changes to benchmark and optimize AI code edit recommenders so they align with natural incremental reasoning rather than static snapshots.
SHIRO achieves geometric mean speedups of 221.5x to 8.8x over four baselines in distributed SpMM on up to 128 GPUs by exploiting sparsity patterns and two-tier network topologies.
PICO is a benchmarking framework for collective operations that decouples portable setup from platform execution, supplies reference MPI implementations, and shows default choices can be up to 5x slower with up to 44% end-to-end training time reductions in simulator replays.
A calibration strategy and uncertainty budget for full two-port S-parameter measurements at mK temperatures over 4-12 GHz using SOLR in coaxial line with SI-traceable room-temperature references and numerical cryo-shift modeling.
Presents a tensor-parallel distributed MPS method with block-cyclic partitioning and pivoted QR that emulates Google's RCS benchmark at bond dimension 16384 on 32 nodes, claiming three orders of magnitude better accuracy than prior methods.
Dynamic replication of predicted overloaded experts in MoE models achieves near-100% GPU utilization and up to 3x faster inference while retaining 90-95% of baseline performance.
CStencil on the WSE-3 achieves up to 342x speedup for 2D stencils versus an adapted single-precision GPU solver and saturates both compute and on-chip memory bandwidth.
HELM adaptively partitions HBM between EMB and KV caches via a three-layer PPO controller and EMB-KV-aware scheduling, reducing P99 latency by 24-38% while achieving 93.5-99.6% SLO satisfaction on production workloads.
A fused gather-GEMM-scatter CUDA kernel achieves 4.6-7.3x end-to-end speedup and 3.2-4.9x lower energy for matrix-free 3D SIMP topology optimization on RTX 4090 compared to three-stage baselines.
A new distributed framework for graph transformer training auto-selects parallel strategies and optimizes sparse operations to deliver up to 6x speedup on 8 GPUs and 78% memory reduction.
A novel MPI-based construction method for spiking neural networks on multi-GPU clusters is introduced, with scaling demonstrated on two cortical models using point-to-point and collective communication.
A clustering-aware correction algorithm using spatial partitioning and projected gradient descent preserves single-linkage clusters in lossy-compressed particle data while keeping competitive compression ratios.
High-resolution power profiles for AI workloads on H100 GPUs are measured and scaled to whole-facility energy demand using a bottom-up model, with the dataset made public.
citing papers explorer
-
Iceberg Beyond the Tip: Co-Compilation of a Quantum Error Detection Code and a Quantum Algorithm
Co-optimization of flexible Iceberg error-detection gadgets with QAOA via tree search improves success probability and post-selection on Quantinuum H2-1 hardware up to 34 algorithmic qubits.
-
NCCLZ: Compression-Enabled GPU Collectives with Decoupled Quantization and Entropy Coding
NCCLZ decouples quantization and entropy coding across NCCL stack layers to enable overlapped compression, delivering up to 9.65x speedup over plain NCCL on scientific and training workloads.
-
Unfolding an Atomistic World: Atomistic Simulation of Reactor Pressure Vessel Steel Across Year-and-Meter Scales
AtomWorld enables the first direct atomistic simulation of RPV steel at year-and-meter scales, handling ten-quintillion-atom systems and simulating one service year in 1.71 days with 92-97% scaling efficiency on leadership supercomputers.
-
Mosaic: Cross-Modal Clustering for Efficient Video Understanding
Mosaic uses cross-modal clusters as the unit for KVCache organization in VLMs to achieve up to 1.38x speedup in streaming long-video understanding.
-
Multi-Agent Orchestration for High-Throughput Materials Screening on a Leadership-Class System
A planner-executor multi-agent system using gpt-oss-120b and Parsl orchestrates scalable high-throughput MOF screening on the Aurora supercomputer with low overhead.
-
NestPipe: Large-Scale Recommendation Training on 1,500+ Accelerators via Nested Pipelining
NestPipe achieves up to 3.06x speedup and 94.07% scaling efficiency on 1,536 workers via dual-buffer inter-batch and frozen-window intra-batch pipelining that overlaps communication with computation.
-
Analyzing Reverse Address Translation Overheads in Multi-GPU Scale-Up Pods
Simulation study shows cold TLB misses in reverse address translation dominate latency for small collectives in multi-GPU pods, causing up to 1.4x degradation, while larger ones see diminishing returns.
-
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
-
Runtime-Orchestrated Second-Order Optimization for Scalable LLM Training
Asteria is a runtime system that enables second-order optimization for LLMs by dynamically distributing optimizer state across GPU, CPU, and NVMe while using asynchronous inverse-root computations and bounded-staleness synchronization.
-
Co-Design Optimization for Data Center Cooling System via Digital Twin
A three-layer co-design optimization using digital twins and surrogate modeling for CDU partitioning and flow control in HPC cooling plants achieves 35.48% annual energy savings, nearly matching the current Frontier design while reducing assignment sensitivity by 93%.
-
CCCL: Node-Spanning GPU Collectives with CXL Memory Pooling
CCCL delivers 1.34-1.94x faster cross-node GPU collectives via CXL memory pooling than 200 Gbps InfiniBand RDMA, with 1.11x LLM training speedup and 2.75x hardware cost reduction.
-
EditFlow: Benchmarking and Optimizing Code Edit Recommendation Systems via Reconstruction of Developer Flows
EditFlow reconstructs temporal developer editing flows from code changes to benchmark and optimize AI code edit recommenders so they align with natural incremental reasoning rather than static snapshots.
-
SHIRO: Near-Optimal Communication Strategies for Distributed Sparse Matrix Multiplication
SHIRO achieves geometric mean speedups of 221.5x to 8.8x over four baselines in distributed SpMM on up to 128 GPUs by exploiting sparsity patterns and two-tier network topologies.
-
PICO: Performance Insights for Collective Operations
PICO is a benchmarking framework for collective operations that decouples portable setup from platform execution, supplies reference MPI implementations, and shows default choices can be up to 5x slower with up to 44% end-to-end training time reductions in simulator replays.
-
Full Two-Port S-Parameters at mK Temperatures: a Calibration Strategy and Uncertainty Budget
A calibration strategy and uncertainty budget for full two-port S-parameter measurements at mK temperatures over 4-12 GHz using SOLR in coaxial line with SI-traceable room-temperature references and numerical cryo-shift modeling.
-
Tensor-Parallel Emulation of Quantum Circuits with Block-Cyclic Distributed Matrix Product States
Presents a tensor-parallel distributed MPS method with block-cyclic partitioning and pivoted QR that emulates Google's RCS benchmark at bond dimension 16384 on 32 nodes, claiming three orders of magnitude better accuracy than prior methods.
-
Fast MoE Inference via Predictive Prefetching and Expert Replication
Dynamic replication of predicted overloaded experts in MoE models achieves near-100% GPU utilization and up to 3x faster inference while retaining 90-95% of baseline performance.
-
Stencil Computations on Cerebras Wafer-Scale Engine
CStencil on the WSE-3 achieves up to 342x speedup for 2D stencils versus an adapted single-precision GPU solver and saturates both compute and on-chip memory bandwidth.
-
One Pool, Two Caches: Adaptive HBM Partitioning for Accelerating Generative Recommender Serving
HELM adaptively partitions HBM between EMB and KV caches via a three-layer PPO controller and EMB-KV-aware scheduling, reducing P99 latency by 24-38% while achieving 93.5-99.6% SLO satisfaction on production workloads.
-
Matrix-Free 3D SIMP Topology Optimization with Fused Gather-GEMM-Scatter Kernels
A fused gather-GEMM-scatter CUDA kernel achieves 4.6-7.3x end-to-end speedup and 3.2-4.9x lower energy for matrix-free 3D SIMP topology optimization on RTX 4090 compared to three-stage baselines.
-
Scalable and Adaptive Parallel Training of Graph Transformer on Large Graphs
A new distributed framework for graph transformer training auto-selects parallel strategies and optimizes sparse operations to deliver up to 6x speedup on 8 GPUs and 78% memory reduction.
-
Scalable Construction of Spiking Neural Networks using up to thousands of GPUs
A novel MPI-based construction method for spiking neural networks on multi-GPU clusters is introduced, with scaling demonstrated on two cortical models using point-to-point and collective communication.
-
Preserving Clusters in Error-Bounded Lossy Compression of Particle Data
A clustering-aware correction algorithm using spatial partitioning and projected gradient descent preserves single-linkage clusters in lossy-compressed particle data while keeping competitive compression ratios.
-
Measurement of Generative AI Workload Power Profiles for Whole-Facility Data Center Infrastructure Planning
High-resolution power profiles for AI workloads on H100 GPUs are measured and scaled to whole-facility energy demand using a bottom-up model, with the dataset made public.
-
Making Room for AI: Multi-GPU Molecular Dynamics with Deep Potentials in GROMACS
GROMACS now runs multi-GPU DeePMD inference for molecular dynamics, reaching 40-66% strong scaling efficiency up to 32 devices on a 15k-atom protein system with over 90% time in inference.
-
Universal Quantum Computer Simulation of 50 Qubits on Europe`s First Exascale Supercomputer Harnessing Its Heterogeneous CPU-GPU Architecture
JUQCS-50 achieves the first 50-qubit universal quantum computer simulation on the JUPITER supercomputer via CPU-GPU memory extension, adaptive encoding, and network optimization, delivering a 16.6-fold speedup over the prior 48-qubit record.
-
RSE of a Quantum Transport Code and its Effects
Experience report on applying RSE practices to libNEGF reveals critical defects and performance issues in a quantum transport code, suggesting such problems are common in Fortran scientific software.
-
Training LLMs on HPC Systems: Best Practices from the OpenGPT-X Project
Engineering report detailing HPC infrastructure, software choices, and performance measurements for training a 7B LLM using 3D parallelism on JUWELS Booster.