SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Measure before modification. Use CrayPat or Intel Advisor to find performance bottlenecks. Roofline analysis is easy with intel advisor. Running from the command line will allow you to submit to the job queue and allow you to analyze results later from the front end. It is possible to do collection only on a subset of ranks.
We evaluate the capabilities of vendor-provided OpenCL implementations for performance portability across multiple computing platforms. The Rodinia benchmark suite is used for this evaluation. We apply the metric defined by Pennycook et al., and we use roofline efficiency from the Roofline performance model as the "performance efficiency" in the metric's definition. We found that the delivered performance portability is similar for several benchmarks, even if the roofline-based performance efficiencies across platforms are very different among the benchmarks. To help distinguish between these instances, we extend the metric by adding the standard deviation of the performance efficiencies for each benchmark. We argue that the standard deviation gives additional insight into performance portability assessment since it adds the performance variability across platforms. Additionally, we discuss the challenges to measure performance portability associated with algorithms and system software. In terms of algorithms, we need to carefully construct the benchmarks and appropriately use the concurrency available on a platform. In terms of system software, we depend on the vendor performance tools to support the desired programming model and runtime to be able to measure the metrics of interest.
GPU accelerators are ubiquitous, but their ecosystem is far less evolved than the host one. Compiler heuristics are often tuned for CPUs and reused for GPU. Similarly, tooling and more evolved optimization techniques are historically not available on GPU targets. In this work, we address one of these shortcomings and enable profile generation and profile-guided optimizations (PGO) for GPU targets. While this is only a single step towards a CPU equivalent ecosystem for offload devices, it shows how old misconceptions on the limitations of GPUs are often not warranted anymore. Through our implementation in LLVM/Offload, we enable device-side PGO for full scientific applications and open up tooling opportunities, including code coverage analysis and compiler-built-in roofline analysis. Our evaluation highlights the performance implications of profile generation, the insights gained from these profiles, and the (missed) opportunities in utilizing the information for GPU compilation.
As quantum computers evolve, simulations of quantum programs on classical computers will be essential in validating quantum algorithms, understanding the effect of system noise, and design applications for future quantum computers. In this paper, we propose a novel multi-GPU programming model called MG-BSP that constructs a virtual BSP machine on top of modern multi-GPU platforms, and tweak the programming model to build a multi-GPU density matrix quantum simulator. We propose and evaluated a new formulation minimizing communication and prove that the transformation conserves original semantics when noise is introduced. We build the tool-chain to support quantum assembly open standard, synthesize testing quantum circuit, and enable ultra-deep quantum simulation. We evaluated our design on four state-of-the-art multi-GPU platforms including the latest DGX-1 and DGX-2 systems. We demonstrate simulation of 1 million gates in 94 minutes, far deeper circuits than has been demonstrated in prior work. A roofline-model analysis show that we have reached near-optimal performance under memory bound.
The scientific computing paradigm has transitioned from compute-intensive to I/O-intensive and memory-intensive in the past decade, especially when data-driven science has become common practice. Numerous empirical I/O-aware scheduling optimizations have been developed by incorporating I/O capacity and bandwidth as constraints into scheduling. Unfortunately, there is a lack of data flow (I/O) characterization tool and an understanding of trade-offs between concurrency, locality, and I/O bandwidth. To bridge the gap, this work 1) presents a set of descriptors to characterize, organize, and visualize I/O profiles, including flow size, I/O bandwidth, and operation count, which group data flows by I/O types, tasks, and files; 2) proposes an I/O Roofline model-based trade-off analysis to find the optimal trade-off between flow operational intensity, concurrency, and flow performance. The I/O descriptors generate useful insights into complicated I/O behaviors, suggesting distinct concurrency, storage, and scheduling to be used by types, tasks, and files. The proposed trade-off analysis guides scheduling decisions that generate resource assignment with the best flow parallelism. We evaluate our I/O-aware scheduling methodology on a highly I/O-intensive workflow–1000 Genomes. The experimental results demonstrate speedups of up to 2.4× compared to the state-of-the- art methods.
We examine a multi-modal approach to educating and training users of an advanced computing technology testbed at the Institute for Advanced Computational Science at Stony Brook University. Ookami provides researchers worldwide with access to 176 Fujitsu A64FX compute nodes, this being the same processor technology powering the Japanese Fugaku supercomputer, the fastest computer in the world since June 2020. However, achieving high-performance on this Arm-based, leadership computing technology requires that users be familiar with details of computer architecture, performance analysis and modeling, and high-performance programming models that are commonly omitted in introductory programming courses. Indeed, regardless of their seniority, many of the testbed users are surprisingly unfamiliar with basic concepts such as vectorization, pipelining, latency/bandwidth, roofline models, computing energy/power, threads, and non-uniform memory access. These same concepts also pervade mainstream x86 technologies, so this is of widespread concern. Due to the national/global nature of our user community that is also very diverse in both discipline and experience, the inability to offer formal classes, and our experience that most people do not tend to read online documentation or training materials in sufficient depth, we have consciously employed multiple approaches that heavily emphasize (online) personal interactions and transfer of skills. Online documentation has been organized around best-practices and FAQs; twice-weekly hackathons and office hours via Zoom enable deep dives by both the team and the user community with multiple broad benefits; a Slack channel provides both real time and archived answers and discussions; and workshops, training and webinars target community needs as they arise. Furthermore, the perspective that these tools are being used in an educational setting rather than just for project communication makes them more effective and contributes to community success.
High-speed chemically active flows pose significant computational challenges due to their disparate space and time scales, with stiff chemistry often dominating simulation time. While modern scientific computing programs achieve exascale performance by leveraging graphics processing units (GPUs), existing GPU-based compressible combustion solvers face critical limitations in memory management, load balancing, and handling the highly localized nature of chemical reactions. To this end, we present a high-performance compressible reacting flow solver built on the AMReX framework and optimized for multi-GPU settings. Here, our approach addresses three GPU performance bottlenecks: memory access patterns through column-major storage optimization, computational workload variability via a bulk-sparse integration strategy for chemical kinetics, and multi-GPU load distribution for adaptive mesh refinement applications. The solver adapts existing matrix-based chemical kinetics formulations to multi-grid contexts. Using representative combustion applications, including 2D and 3D detonations and a 3D jet-in-crossflow configuration, we demonstrate 1.4–5× performance improvements over initial implementations on an in-house cluster of NVIDIA H100 GPUs, and near-ideal weak scaling on the Frontier supercomputer (Oak Ridge Leadership Computing Facility) with up to 1024 AMD Instinct MI250X GPUs. Roofline analysis reveals substantial improvements in arithmetic intensity for both convection (∼ 10 ×) and chemistry (∼ 4 ×) routines, confirming efficient utilization of GPU memory bandwidth and computational resources.
Tensor computations present significant performance chal- lenges that impact a wide spectrum of applications ranging from machine learning, healthcare analytics, social network analysis, data mining to quantum chemistry and signal processing. Efforts to improve the perfor- mance of tensor computations include exploring data layout, execution scheduling, and parallelism in common tensor kernels. This work presents a benchmark suite for arbitrary-order sparse tensor kernels using state- of-the-art tensor formats: coordinate (COO) and hierarchical coordinate (HiCOO) on CPUs and GPUs. It presents a set of reference tensor kernel implementations that are compatible with real-world tensors and power law tensors extended from synthetic graph generation techniques. We also propose Roofline performance models for these kernels to provide insights of computer platforms from sparse tensor view. This benchmark suite along with the synthetic tensor generator is publicly available.
Pairwise sequence alignment is one of the most computationally intensive kernels in genomic data analysis, accounting for more than 90% of the runtime for key bioinformatics applications. This method is particularly expensive for third-generation sequences due to the high computational cost of analyzing sequences of length between 1Kb and 1Mb. Given the quadratic overhead of exact pairwise algorithms for long alignments, the community primarily relies on approximate algorithms that search only for high-quality alignments and stop early when one is not found. In this work, we present the first GPU optimization of the popular X-drop alignment algorithm, that we named LOGAN. Results show that our high-performance multi-GPU implementation achieves up to 181.6 GCUPS and speed-ups up to 6.6× and 30.7× using 1 and 6 NVIDIA Tesla V100, respectively, over the state-of-the-art software running on two IBM Power9 processors using 168 CPU threads, with equivalent accuracy. We also demonstrate a 2.3× LOGAN speed-up versus ksw2, a state-of-art vectorized algorithm for sequence alignment implemented in minimap2, a long-read mapping software. Furthermore, to highlight the impact of our work on a real-world application, we couple LOGAN with a many-to-many long-read alignment software called BELLA, and demonstrate that our implementation improves the overall BELLA runtime by up to 10.6×. Finally, we adapt the Roofline model for LOGAN and demonstrate that our implementation is near optimal on the NVIDIA Tesla V100s.