Search NASA⌕ Search

SEARCH · Search NASA

Results for “Threading”

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.

At least 721 records · Page 40

Multipoint Aerostructural Optimization of Wind Turbine Rotors Using a Coupled Blade‐Resolved Aerostructural Solver

Physics‐based design optimization workflows thread the needle between computational cost limitations and simulation complexity, often compromising between modeling detail and the range of operating design conditions. Multipoint aerostructural optimization of wind turbine rotors has so far been confined to low‐fidelity analyses or to high‐fidelity studies with simplified structural models, leaving the most complex design trade‐offs unexplored. We close this gap by performing the first tightly coupled gradient‐based multipoint aerostructural rotor optimization using 3D aerodynamic and structural solvers with discrete coupled adjoints. The optimizer simultaneously varies blade planform, airfoil shapes, and structural thickness through more than 270 design variables, minimizing a weighted combination of rotor mass and power across multiple wind speeds. Applied to a modified DTU 10‐MW benchmark under conservative structural and aerodynamic constraints, our multipoint optimization reduces rotor mass by up to 36% and increases power by 12%–15% across the main operating conditions; biasing the objective toward power yields power gains up to 18% and a 17% mass reduction. For a nominal wind distribution, 3‐point rotor designs accounting for low RPM and high thrust conditions capture dominant trade‐offs and outperform single‐point designs. Adding two off‐design points changes individual‐condition power by less than 3% but leaves the weighted average within 0.5%, and the mass‐power bias has a stronger effect on the final design than the operating‐point weighting itself. Our framework extends naturally to richer load cases and site‐specific wind distributions, providing a basis for high‐fidelity multipoint design earlier in industrial workflows.

17 WIND ENERGY↗

Compile-Time Library Call Detection Using CAASCADE and XALT

CAASCADE — Compiler-Assisted Application Source Code Analysis and DatabasE—is a tool that summarizes the use of parallel programming language features in application source code using compiler technology. This paper discusses the library detection capability within CAASCADE to find information about the usage of scientific libraries within the source code. The information that CAASCADE collects provides insights into the usage of library calls in an applications. CAASCADE can classify the APIs by scientific libraries (e.g. LAPACK, BLAS, FFTW, etc). It can also detect the context in which a library API is being invoked, for example within a serial or multi-threaded region. To collect this information, CAASCADE uses compiler plugins that summarize procedural information and uses Apache Spark to do inter-procedural analysis to reconstruct call chains. In addition to this, we also integrated CAASCADE to work with XALT to collect library information based on linkage and modules installed on a system.

Zhao, Jisheng↗

Benchmarking Operators in Deep Neural Networks for Improving Performance Portability of SYCL

SYCL is a portable programming model for heterogeneous computing, so it is important to obtain reasonable performance portability of SYCL. Towards the goal of better understanding and improving performance portability of SYCL for machine learning workloads, we have been developing benchmarks for basic operators in deep neural networks (DNNs). These operators could be offloaded to heterogeneous computing devices such as graphics processing units (GPUs) to speed up computation. In this paper, we introduce the benchmarks, evaluate the performance of the operators on GPU-based systems, and describe the causes of the performance gap between the SYCL and Compute Unified Device Architecture (CUDA) kernels. We find that the causes are related to the utilization of the texture cache for read-only data, optimization of the memory accesses with strength reduction, use of local memory, and register usage per thread. We hope that the efforts of developing benchmarks for studying performance portability will stimulate discussion and interactions within the community.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗

A Flexible CUDA LU-based Solver for Small, Batched Linear Systems

This chapter presents the implementation of a batched CUDA solver based on LU factorization for small linear systems. This solver may be used in applications such as reactive flow transport models, which apply the Newton-Raphson technique to linearize and iteratively solve the sets of non linear equations that represent the reactions for ten of thousands to millions of physical locations. The implementation exploits somewhat counterintuitive GPGPU programming techniques: it assigns the solution of a matrix (representing a system) to a single CUDA thread, does not exploit shared memory and employs dynamic memory allocation on the GPUs. These techniques enable our implementation to simultaneously solve sets of systems with over 100 equations and to employ LU decomposition with complete pivoting, providing the higher numerical accuracy required by certain applications. Other currently available solutions for batched linear solvers are limited by size and only support partial pivoting, although they may result faster in certain conditions. We discuss the code of our implementation and present a comparison with the other implementations, discussing the various tradeoffs in terms of performance and flexibility. This work will enable developers that need batched linear solvers to choose whichever implementation is more appropriate to the features and the requirements of their applications, and even to implement dynamic switching approaches that can choose the best implementation depending on the input data.

bached linear solvers↗

Optimizing Excited-State Electronic-Structure Codes for Intel Knights Landing: A Case Study on the BerkeleyGW Software

We profile and optimize calculations performed with the BerkeleyGW code on the Xeon-Phi architecture. BerkeleyGW depends both on hand-tuned critical kernels as well as on BLAS and FFT libraries. We describe the optimization process and performance improvements achieved. We discuss a layered parallelization strategy to take advantage of vector, thread and node-level parallelism. We discuss locality changes (including the consequence of the lack of L3 cache) and effective use of the on-package high-bandwidth memory. We show preliminary results on Knights-Landing including a roofline study of code performance before and after a number of optimizations. We find that the GW method is particularly well-suited for many-core architectures due to the ability to exploit a large amount of parallelism over plane-wave components, band-pairs, and frequencies.

97 MATHEMATICS AND COMPUTING↗

On the Scalability of Data Reduction Techniques in Current and Upcoming HPC Systems from an Application Perspective

We implement and benchmark parallel I/O methods for the fully-manycore driven particle-in-cell code PIConGPU. Identifying throughput and overall I/O size as a major challenge for applications on today’s and future HPC systems, we present a scaling law characterizing performance bottlenecks in state-of-the-art approaches for data reduction. Consequently, we propose, implement and verify multi-threaded data-transformations for the I/O library ADIOS as a feasible way to trade underutilized host-side compute potential on heterogeneous systems for reduced I/O latency.

Huebl, Axel↗

Design Principles for Sparse Matrix Multiplication on the GPU

We implement two novel algorithms for sparse-matrix dense-matrix multiplication (SpMM) on the GPU. Our algorithms expect the sparse input in the popular compressed-sparse-row (CSR) format and thus do not require expensive format conversion. While previous SpMM work concentrates on thread-level parallelism, we additionally focus on latency hiding with instruction-level parallelism and load-balancing. We show, both theoretically and experimentally, that the proposed SpMM is a better fit for the GPU than previous approaches. We identify a key memory access pattern that allows efficient access into both input and output matrices that is crucial to getting excellent performance on SpMM. By combining these two ingredients---(i) merge-based load-balancing and (ii) row-major coalesced memory access---we demonstrate a 4.1x peak speedup and a 31.7% geomean speedup over state-of-the-art SpMM implementations on real-world datasets.

97 MATHEMATICS AND COMPUTING↗

CLOMP: Accurately Characterizing OpenMP Application Overheads

Despite its ease of use, OpenMP has failed to gain widespread use on large scale systems, largely due to its failure to deliver sufficient performance. Our experience indicates that the cost of initiating OpenMP regions is simply too high for the desired OpenMP usage scenario of many applications. In this paper, we introduce CLOMP, a new benchmark to characterize this aspect of OpenMP implementations accurately. CLOMP complements the existing EPCC benchmark suite to provide simple, easy to understand measurements of OpenMP overheads in the context of application usage scenarios. Our results for several OpenMP implementations demonstrate that CLOMP identifies the amount of work required to compensate for the overheads observed with EPCC.We also show that CLOMP also captures limitations for OpenMP parallelization on SMT and NUMA systems. Finally, CLOMPI, our MPI extension of CLOMP, demonstrates which aspects of OpenMP interact poorly with MPI when MPI helper threads cannot run on the NIC.

99 GENERAL AND MISCELLANEOUS↗

A Case for Including Transactions in OpenMP

Transactional Memory (TM) has received significant attention recently as a mechanism to reduce the complexity of shared memory programming. We explore the potential of TM to improve OpenMP applications. We combine a software TM (STM) system to support transactions with an OpenMP implementation to start thread teams and provide task and loop-level parallelization. We apply this system to two application scenarios that reflect realistic TM use cases. Our results with this system demonstrate that even with the relatively high overheads of STM, transactions can outperform OpenMP critical sections by 10%. Overall, our study demonstrates that extending OpenMP to include transactions would ease programming effort while allowing improved performance.

99 GENERAL AND MISCELLANEOUS↗

A Proposal for User-defined Reductions in OpenMP

Reductions are commonly used in parallel programs to produce a global result from partial results computed in parallel. Currently, OpenMP only supports reductions for primitive data types and a limited set of base language operators. This is a significant limitation for those applications that employ user-defined data types (e. g., objects). Implementing manual reduction algorithms makes software development more complex and error-prone. Additionally, an OpenMP runtime system cannot optimize a manual reduction algorithm in ways typically applied to reductions on primitive types. In this paper, we propose new mechanisms to allow the use of most pre-existing binary functions on user-defined data types as User-Defined Reduction (UDR) operators. Our measurements show that our UDR prototype implementation provides consistently good performance across a range of thread counts without increasing general runtime overheads.

97 MATHEMATICS AND COMPUTING↗

On the Performance of an Algebraic Multigrid Solver on Multicore Clusters

Algebraic multigrid (AMG) solvers have proven to be extremely efficient on distributed-memory architectures. However, when executed on modern multicore cluster architectures, we face new challenges that can significantly harm AMG's performance. We discuss our experiences on such an architecture and present a set of techniques that help users to overcome the associated problems, including thread and process pinning and correct memory associations. We have implemented most of the techniques in a MultiCore SUPport library (MCSup), which helps to map OpenMP applications to multicore machines. We present results using both an MPI-only and a hybrid MPI/OpenMP model.

99 GENERAL AND MISCELLANEOUS↗

Experiences with High-Level Programming Directives for Porting Applications to GPUs. In: Facing the Multicore--Challenge II, Lecture Notes in Computer Science.

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the different node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressivity and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

CUDA Implementation↗

Holographic entanglement distillation from the surface state correspondence

We study correlations between geometric subfactors living on the Ryu-Takayanagi surface that bounds the entanglement wedge. Using the surface-state correspondence and the bit threads program, we are able to calculate mutual information and conditional mutual information between subfactors. This enables us to count the shared Bell pairs between subfactors, and we propose an entanglement distillation procedure over these subsystems via a SWAP gate protocol. We comment on extending to multipartite entanglement.

99 GENERAL AND MISCELLANEOUS↗

Monopole operators in U(1) Chern-Simons-matter theories

We report on monopole operators at the infrared fixed points of U(1) Chern-Simons-matter theories (QED 3 , scalar QED 3 , $\mathcal{N}$=1 SQED 3 , and $\mathcal{N}$=2 SQED 3 ) with N matter flavors and Chern-Simons level k. We work in the limit where both N and k are taken to be large with κ = k/ N fixed. In this limit, we extract information about the low-lying spectrum of monopole operators from evaluating the S 2 × S 1 partition function in the sector where the S 2 is threaded by magnetic flux 4πq. At leading order in N , we find a large number of monopole operators with equal scaling dimensions and a wide range of spins and flavor symmetry irreducible representations. In two simple cases, we deduce how the degeneracy in the scaling dimensions is broken by the 1/ N corrections. For QED 3 at κ = 0, we provide conformal bootstrap evidence that this near-degeneracy is in fact maintained to small values of N . For $\mathcal{N}$=2 SQED 3 , we discover that the lowest dimension monopole operator is generically non-BPS.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

blocks_3d: software for general 3d conformal blocks

We introduce the software blocks_3d for computing four-point conformal blocks of operators with arbitrary Lorentz representations in 3d CFTs. It uses Zamolodchikov-like recursion relations to numerically compute derivatives of blocks around a crossing-symmetric configuration. It is implemented as a heavily optimized, multi-threaded, C++ application. We give performance benchmarks for correlators containing scalars, fermions, and stress tensors. As an example application, we recompute bootstrap bounds on four-point functions of fermions and study whether a previously observed sharp jump can be explained using the “fake primary” effect. We conclude that the fake primary effect cannot fully explain the jump and the possible existence of a “dead-end” CFT near the jump merits further study.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The metamorphosis of semi-classical mechanisms of confinement: from monopoles on ℝ 3 × S 1 to center-vortices on ℝ 2 × T 2

There are two distinct regimes of Yang-Mills theory where we can demonstrate confinement, the existence of a mass gap, and the multi-branch structure of the effective potential as a function of the theta angle using a reliable semi-classical calculation. The two regimes are deformed Yang-Mills theory on ℝ 3 × S 1 , and Yang-Mills theory on ℝ 2 × T 2 where the torus is threaded by a ’t Hooft flux. The weak coupling regime is ensured by the small size of the circle or torus. In the first case the confinement mechanism is related to self-dual monopoles, whereas in the second case self-dual center-vortices play a crucial role. These two topological objects are distinct. In particular, they have different mutual statistics with Wilson loops. On the other hand, they carry the same topological charge and action. We consider the theory on ℝ × T 2 × S 1 and extrapolate both the monopole and vortex regimes to a quantum mechanical domain, where a cross-over takes place. Both sides of the cross-over are described by a deformed ℤ N TQFT. On ℝ 2 × S 1 × S 1 , we derive an effective field theory (EFT) of vortices from the EFT of monopoles in the presence of a ’t Hooft flux. This construction is based on a two-stage Higgs mechanism, reducing SU(N) to U(1) N−1 in 3d first, followed by reduction to a ℤ N EFT in 2d in the second step. This result shows how monopoles transmute into center-vortices, and suggests adiabatic continuity between the two confinement mechanisms. The basic mechanism is flux fractionalization: the magnetic flux of the monopoles splits up and is collimated in such a way that 2d Wilson loops detect it as a center vortex.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Reflected entropy in random tensor networks. Part III. Triway cuts

For general random tensor network states at large bond dimension, we prove that the integer Rényi reflected entropies (away from phase transitions) are determined by minimal triway cuts through the network. This generalizes the minimal cut description of bipartite entanglement for these states. A natural extrapolation away from integer Rényi parameters, suggested by the triway cut problem, implies the holographic conjecture S R = 2EW, where S R is the reflected entropy and EW is the entanglement wedge cross-section. Minimal triway cuts can be formulated as integer programs which cannot be relaxed to find a dual maximal flow/bit-thread description. This sheds light on the gap between the existence of tripartite entanglement in holographic states and the bipartite entanglement structure motivated by bit-threads. In particular, we prove that the Markov gap that measures tripartite entanglement is lower bounded by the integrality gap of the integer program that computes the triway cut.

AdS-CFT correspondence↗

Towards High-Speed friction stir welding of 25 mm Thick AA2139-T8: tool innovation and process development

Joining thick plates (≥ 12 mm) of aluminum (Al) alloys is challenging due to high tool forces, uneven material flow, and non-uniform heat distribution through the material’s thickness. Recent advancements in tool design and welding parameters for friction stir welding (FSW) in thick plates butt joining have encountered a developmental plateau, highlighting the need for innovative approaches to overcome existing limitations. Here, this study focuses on the systematic development of single-pass and double-pass FSW processes for high-strength aluminum alloy AA2139-T8, to improve joint efficiency and enable high-speed welding capabilities. Experimental evidence is presented for a novel tool design with opposing pin threads, enabling high-speed (178 mm/min) single-pass friction stir butt welding of 25 mm thick AA2139-T8. A series of FSW trials was conducted both in air and with a trailing water spray, using steel backing plates (BPs) to investigate the impact of quenching and cooling rates on process response and joint performance. A joint efficiency of 83% was attained using the novel tool features and effectively controlling process forces and thermal boundary conditions.

AA2139-T8↗