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 1,225 records · Page 68

INTEGRATION OF PANDA WORKLOAD MANAGEMENT SYSTEM WITH SUPERCOMPUTERS

Abstract The Large Hadron Collider (LHC), operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. Experiments at the LHC explore the funda- mental nature of matter and the basic forces that shape our universe, and were recently credited for the dis- covery of a Higgs boson. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 140 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data cen- ters are physically scattered all over the world. While PanDA currently uses more than 250000 cores with a peak performance of 0.3+ petaFLOPS, next LHC data taking runs will require more resources than Grid computing can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, Europe and Russia (in particular with Titan supercomputer at Oak Ridge Leadership Com- puting Facility (OLCF), Supercomputer at the National Research Center Kurchatov Institute , IT4 in Ostrava, and others). The current approach utilizes a modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run single- threaded workloads in parallel on Titan s multi-core worker nodes. This implementation was tested with a variety of Monte-Carlo workloads on several supercomputing platforms. We will present our current accom- plishments in running PanDA WMS at supercomputers and demonstrate our ability to use PanDA as a portal independent of the computing facility s infrastructure for High Energy and Nuclear Physics, as well as other data-intensive science applications, such as bioinformatics and astro-particle physics.

De, K↗

Highly scalable distributed-memory sparse triangular solution algorithms.

This paper presents a highly efficient distributed-memory parallel sparse triangular solver. The triangular solution phase is often performed following factorization phase in the sparse linear solvers and has become increasingly computa­tionally expensive for direct solvers with many right hand sides (RHSs) or preconditioned iterative solvers. However, the low arithmetic intensity and sequential nature of the triangular solve algorithm pose performance challenges for its large-scale distributed-memory parallelization. In this work, we propose several strategies to enhance scalability of an algorithm with 2D block cyclic process layout. First, an asynchronous binary-tree-based communication scheme implemented via non-blocking MPI functions is leveraged to broadcast partial solutions and reduce partial updates among a subset of processes for each block column and row of the triangular matrix, respectively. This scheme reduces message latency, improves communication load balance and significantly accelerates asynchronous execution of the triangular solve. In addition, efficient BLAS operations and threading implementations are exploited to accelerate local computations and further reduce process idle time. The proposed strategies are implemented in SuperLU_DIST and numerical experiments show up to 4.4x improvement with one right-hand side and up to 6.1x improvement with 50 right- hand sides on 4096 processes, compared to the current release. This is the first time that sparse triangular solution is demonstrated strong scaling on more than 4000 cores.

Liu, Yang↗

Software for Sparse Tensor Decomposition on Emerging Computing Architectures

In this paper, we develop software for decomposing sparse tensors that is portable to and performant on a variety of multicore, manycore, and GPU computing architectures. The result is a single code whose performance matches optimized architecture-specific implementations. The key to a portable approach is to determine multiple levels of parallelism that can be mapped in different ways to different architectures, and we explain how to do this for the matricized tensor times Khatri--Rao product (MTTKRP), which is the key kernel in canonical polyadic tensor decomposition. Our implementation leverages the Kokkos framework, which enables a single code to achieve high performance across multiple architectures that differ in how they approach fine-grained parallelism. We also introduce a new construct for portable thread-local arrays, which we call compile-time polymorphic arrays. Not only are the specifics of our approaches and implementation interesting for tuning tensor computations, but they also provide a roadmap for developing other portable high-performance codes. As a last step in optimizing performance, we modify the MTTKRP algorithm itself to do a permuted traversal of tensor nonzeros to reduce atomic-write contention. Lastly, we test the performance of our implementation on 16- and 68-core Intel CPUs and the K80 and P100 NVIDIA GPUs, showing that we are competitive with state-of-the-art architecture-specific codes while having the advantage of being able to run on a variety of architectures.

97 MATHEMATICS AND COMPUTING↗

FROSch Preconditioners for Land Ice Simulations of Greenland and Antarctica

Numerical simulations of Greenland and Antarctic ice sheets involve the solution of large-scale highly nonlinear systems of equations on complex shallow geometries. This work is concerned with the construction of Schwarz preconditioners for the solution of the associated tangent problems, which are challenging for solvers mainly because of the strong anisotropy of the meshes and wildly changing boundary conditions that can lead to poorly constrained problems on large portions of the domain. In this study, two-level generalized Dryja--Smith--Widlund (GDSW)--type Schwarz preconditioners are applied to different land ice problems, i.e., a velocity problem, a temperature problem, as well as the coupling of the former two problems. We employ the message passing interface (MPI)--parallel implementation of multilevel Schwarz preconditioners provided by the package FROSch (fast and robust Schwarz) from the Trilinos library. The strength of the proposed preconditioner is that it yields out-of-the-box scalable and robust preconditioners for the single physics problems. To the best of our knowledge, this is the first time two-level Schwarz preconditioners have been applied to the ice sheet problem and a scalable preconditioner has been used for the coupled problem. The preconditioner for the coupled problem differs from previous monolithic GDSW preconditioners in the sense that decoupled extension operators are used to compute the values in the interior of the subdomains. Additionally, several approaches for improving the performance, such as reuse strategies and shared memory OpenMP parallelization, are explored. In our numerical study we target both uniform meshes of varying resolution for the Antarctic ice sheet as well as nonuniform meshes for the Greenland ice sheet. We present several weak and strong scaling studies confirming the robustness of the approach and the parallel scalability of the FROSch implementation. Among the highlights of the numerical results are a weak scaling study for up to 32,K processor cores (8,K MPI ranks and 4 OpenMP threads) and 566,M degrees of freedom for the velocity problem as well as a strong scaling study for up to 4,K processor cores (and MPI ranks) and 68,M degrees of freedom for the coupled problem.

97 MATHEMATICS AND COMPUTING↗

A Performance Portable, Fully Implicit Landau Collision Operator with Batched Linear Solvers

Modern accelerators use hierarchical parallel programming models that enable massive multithreading within a processing element (PE), with multiple PEs per device driven by traditional processes. Batching is a technique for exposing PE-level parallelism in algorithms that have traditionally run on MPI processes or multiple threads within a single process. Opportunities for batching arise in, for example, kinetic discretizations of magnetized plasmas where collisions are advanced in velocity space at each spatial point independently. This paper builds on previous work on a high-performance, fully nonlinear, Landau collision operator by batching the linear solver, as well as batching the spatial point problems and adding new support for multiple grids for multiscale, multispecies problems. An anisotropic relaxation verification test that agrees well with previously published results and analytical models is presented. The performance results from NVIDIA A100 and AMD MI250X nodes are presented with hardware utilization analysis for each architecture. Finally, the entire implicit Landau operator time advance is implemented in Kokkos for performance portability, running entirely on the device and is available in the PETSc numerical library.

97 MATHEMATICS AND COMPUTING↗

Performance of an optical TPC Geant4 simulation with opticks GPU-accelerated photon propagation

We investigate the performance of Opticks, a NVIDIA OptiX API 7.5 GPU-accelerated photon propagation tool compared with a single-threaded Geant4 simulation. We compare the simulations using an improved model of the NEXT-CRAB-0 gaseous time projection chamber. Performance results suggest that Opticks improves simulation speeds by between 58.47 ± 0.02 and 181.39 ± 0.28 times relative to a CPU-only Geant4 simulation and these results vary between different types of GPU and CPU. A detailed comparison shows that the number of detected photons, along with their times and wavelengths, are in good agreement between Opticks and Geant4.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Making Data Prefetch Smarter: Adaptive Prefetching on POWER7

Hardware data prefetch engines are integral parts of many general purpose server-class microprocessors in the eld to- day. Some prefetch engines allow the user to change some of their parameters. The prefetcher, however, is usually enabled in a default conguration during system bring-up and dynamic reconguration of the prefetch engine is not an autonomic feature of current machines. Conceptually, however, it is easy to infer that commonly used prefetch algorithms, when applied in a xed mode will not help per- formance in many cases. In fact, they may actually de- grade performance due to useless bus bandwidth consump- tion and cache pollution. In this paper, we present an adap- tive prefetch scheme that dynamically modies the prefetch settings in order to adapt to the workload requirements. We implement and evaluate adaptive prefetching in the con- text of an existing, commercial processor, namely the IBM POWER7. Our adaptive prefetch mechanism improves per- formance with respect to the default prefetch setting up to 2.7X and 30% for single-threaded and multiprogrammed workloads, respectively. Categories

Jimenez, Victor↗

Steal Tree: Low-Overhead Tracing of Work Stealing Schedulers

Work stealing is a popular approach to scheduling task-parallel programs. The flexibility inherent in work stealing when dealing with load imbalance results in seemingly irregular computation structures, complicating the study of its runtime behavior. In this paper, we present an approach to efficiently trace async-finish parallel programs scheduled using work stealing. We identify key properties that allow us to trace the execution of tasks with low time and space overheads. We also study the usefulness of the proposed schemes in supporting algorithms for data-race detection and retentive steal- ing presented in the literature. We demonstrate that the perturbation due to tracing is within the variation in the execution time with 99% confidence and the traces are concise, amounting to a few tens of kilobytes per thread in most cases. We also demonstrate that the traces enable significant reductions in the cost of detecting data races and result in low, stable space overheads in supporting retentive stealing for async-finish programs.

async-finish parallelism↗

Diagnosing the Causes and Severity of One-sided Message Contention

Two trends suggest network contention for one-sided messages is poised to become a performance problem that concerns application developers: an increased interest in one-sided programming models and a rising ratio of hardware threads to network injection bandwidth. Unfortunately, it is difficult to reason about network contention and one-sided messages because one-sided tasks can either decrease or increase contention. We present effective and portable techniques for diagnosing the causes and severity of one-sided message contention. To detect that a message is affected by contention, we maintain statistics representing instantaneous (non-local) network resource demand. Using lightweight measurement and modeling, we identify the portion of a message's latency that is due to contention and whether contention occurs at the initiator or target. We attribute these metrics to program statements in their full static and dynamic context. We characterize contention for an important computational chemistry benchmark on InfiniBand, Cray Aries, and IBM Blue Gene/Q interconnects. We pinpoint the sources of contention, estimate their severity, and show that when message delivery time deviates from an ideal model, there are other messages contending for the same network links. With a small change to the benchmark, we reduce contention up to 50% and improve total runtime as much as 20%.

Network Contention, Performance Analysis, Dynamic ↗

Exploiting communication concurrency on high performance computing systems

Although logically available, applications may not exploit enough instantaneous communication concurrency to maximize hardware utilization on HPC systems. This is exacerbated in hybrid programming models such as SPMD+OpenMP. We present the design of a "multi-threaded" runtime able to transparently increase the instantaneous network concurrency and to provide near saturation bandwidth, independent of the application configuration and dynamic behavior. The runtime forwards communication requests from application level tasks to multiple communication servers. Our techniques alleviate the need for spatial and temporal application level message concurrency optimizations. Experimental results show improved message throughput and bandwidth by as much as 150% for 4KB bytes messages on InfiniBand and by as much as 120% for 4KB byte messages on Cray Aries. For more complex operations such as all-to-all collectives, we observe as much as 30% speedup. This translates into 23% speedup on 12,288 cores for a NAS FT implemented using FFTW. We also observe as much as 76% speedup on 1,500 cores for an already optimized UPC+OpenMP geometric multigrid application using hybrid parallelism.

97 MATHEMATICS AND COMPUTING↗

Parallel implementation and performance optimization of the configuration-interaction method

The configuration-interaction (CI) method, long a popular approach to describe quantum many-body systems, is cast as a very large sparse matrix eigenpair problem with matrices whose dimension can exceed one billion. Such formulations place high demands on memory capacity and memory bandwidth - - two quantities at a premium today. In this paper, we describe an efficient, scalable implementation, BIGSTICK, which, by factorizing both the basis and the interaction into two levels, can reconstruct the nonzero matrix elements on the fly, reduce the memory requirements by one or two orders of magnitude, and enable researchers to trade reduced resources for increased computational time. We optimize BIGSTICK on two leading HPC platforms - - the Cray XC30 and the IBM Blue Gene/Q. Specifically, we not only develop an empirically-driven load balancing strategy that can evenly distribute the matrix-vector multiplication across 256K threads, we also developed techniques that improve the performance of the Lanczos reorthogonalization. Combined, these optimizations improved performance by 1.3-8× depending on platform and configuration.

97 MATHEMATICS AND COMPUTING↗

Enabling the High Level Synthesis of Data Analytics Accelerators

Conventional High Level Synthesis (HLS) tools mainly tar- get compute intensive kernels typical of digital signal pro- cessing applications. We are developing techniques and ar- chitectural templates to enable HLS of data analytics appli- cations. These applications are memory intensive, present fine-grained, unpredictable data accesses, and irregular, dy- namic task parallelism. We discuss an architectural tem- plate based around a distributed controller to efficiently ex- ploit thread level parallelism. We present a memory in- terface that supports parallel memory subsystems and en- ables implementing atomic memory operations. We intro- duce a dynamic task scheduling approach to efficiently ex- ecute heavily unbalanced workload. The templates are val- idated by synthesizing queries from the Lehigh University Benchmark (LUBM), a well know SPARQL benchmark.

Data Analytics↗

Galactos: computing the anisotropic 3-point correlation function for 2 billion galaxies

The nature of dark energy and the complete theory of gravity are two central questions currently facing cosmology. A vital tool for addressing them is the 3-point correlation function (3PCF), which probes deviations from a spatially random distribution of galaxies. However, the 3PCF's formidable computational expense has prevented its application to astronomical surveys comprising millions to billions of galaxies. We present Galactos, a high-performance implementation of a novel, O( N 2 ) algorithm that uses a load-balanced k-d tree and spherical harmonic expansions to compute the anisotropic 3PCF. Our implementation is optimized for the Intel Xeon Phi architecture, exploiting SIMD parallelism, instruction and thread concurrency, and significant L1 and L2 cache reuse, reaching 39% of peak performance on a single node. Galactos scales to the full Cori system, achieving 9.8 PF (peak) and 5.06 PF (sustained) across 9636 nodes, making the 3PCF easily computable for all galaxies in the observable universe.

79 ASTRONOMY AND ASTROPHYSICS↗

Cache Locality Optimization for Recursive Programs

We present an approach to optimize the cache locality for recursive programs by dynamically splicing--recursively interleaving--the execution of distinct function invocations. By utilizing data effect annotations, we identify concurrency and data reuse opportunities across function invocations and interleave them to reduce reuse distance. We present algorithms that efficiently track effects in recursive programs, detect interference and dependencies, and interleave execution of function invocations using user-level (non-kernel) lightweight threads. To enable multi-core execution, a program is parallelized using a nested fork/join programming model. Our cache optimization strategy is designed to work in the context of a random work stealing scheduler. We present an implementation using the MIT Cilk framework that demonstrates significant improvements in sequential and parallel performance, competitive with a state-of-the-art compile-time optimizer for loop programs and a domain- specific optimizer for stencil programs.

Lifflander, Jonathan↗

Juggler: a dependence-aware task-based execution framework for GPUs

Scientific applications with single instruction, multiple data (SIMD) computations show considerable performance improvements when run on today's graphics processing units (GPUs). However, the existence of data dependences across thread blocks may significantly impact the speedup by requiring global synchronization across multiprocessors (SMs) inside the GPU. To efficiently run applications with interblock data dependences, we need fine-granular task-based execution models that will treat SMs inside a GPU as stand-alone parallel processing units. Such a scheme will enable faster execution by utilizing all internal computation elements inside the GPU and eliminating unnecessary waits during device-wide global barriers.In this paper, we propose Juggler, a task-based execution scheme for GPU workloads with data dependences. The Juggler framework takes applications embedding OpenMP 4.5 tasks as input and executes them on the GPU via an efficient in-device runtime, hence eliminating the need for kernel-wide global synchronization. Juggler requires no or little modification to the source code, and once launched, the runtime entirely runs on the GPU without relying on the host through the entire execution. We have evaluated Juggler on an NVIDIA Tesla P100 GPU and obtained up to 31% performance improvement against global barrier based implementation, with minimal runtime overhead.

Belviranli, Mehmet E.↗

Optimizing Tensor Contractions in CCSD(T) for Efficient Execution on GPUs

Tensor contractions are higher dimensional analogs of matrix multiplications, used in many computational contexts such as high order models in quantum chemistry, deep learning, nite element methods etc. In contrast to the wide availability of high-performance libraries for matrix multiplication on GPUs, the same is not true for tensor contractions. In this paper, we address the optimization of a set of symmetrized tensor contractions that form the computational bottleneck in the CCSD(T) coupled-cluster method in computational chemistry suites like NWChem. Some of the challenges in optimizing tensor contractions that arise in practice from the variety of dimensionalities and shapes for tensors include effective mapping of the high-dimensional iteration space to threads, choice of data buffering in shared-memory and registers, and tile sizes for multi-level tiling. Furthermore, in the case of symmetrized tensor contractions in CCSD(T), it is also a challenge to fuse contractions to reduce data movement cost by exploiting reuse of intermediate tensors. In this paper, we developed an efficient GPU implementation of the tensor contractions in CCSD(T) using shared-memory buffering, register tiling, loop fusion and register transpose. Experimental results demonstrate significant improvement over the current state-of-the-art.

Kim, Jinsung↗

High-Performance Sparse Matrix-Matrix Products on Intel KNL and Multicore Architectures

Sparse matrix-matrix multiplication (SpGEMM) is a computational primitive that is vastly used in areas ranging from traditional numerical applications to recent big data analysis and machine learning. While many SpGEMM algorithms have been proposed, hardware specific optimizations for multi- and many-core processors are lacking and a detailed analysis of their performance under various use cases and matrices is not available. We firstly identify and mitigate multiple bottlenecks with memory management and thread scheduling on Intel Xeon Phi (Knights Landing or KNL). Specifically targeting multi- and many-core processors, we develop a hash-table-based algorithm and optimize a heap-based shared-memory SpGEMM algorithm. We investigate their performance together with other publicly available codes. Different from the literature, our evaluation also includes use cases that are representative of real graph algorithms, such as multi-source breadth-first search or triangle counting. Our hash-table and heap-based algorithms are showing significant speedups from libraries in the majority of the cases while different algorithms dominate the other scenarios with different matrix size, sparsity, compression factor and operation type. We wrap up in-depth evaluation results and make a recipe to give the best SpGEMM algorithm for target scenario. In conclusion, a critical finding is that hash-table-based SpGEMM gets a significant performance boost if the nonzeros are not required to be sorted within each row of the output matrix.

97 MATHEMATICS AND COMPUTING↗

System Architecture Directions for Post-SoC/32-bit Networked Sensors

The emergence of low-power 32-bit Systems-on-Chip (SoCs), which integrate a 32-bit MCU, radio, and flash, presents an opportunity to re-examine design points and trade-offs at all levels of the system architecture of networked sensors. To this end, we develop a post-SoC/32-bit design point called Hamilton, showing that using integrated components enables a ~$7 core and shifts hardware modularity to design time. We study the interaction between hardware and embedded operating systems, identifying that (1) post-SoC motes provide lower idle current (5.9 μA) than traditional 16-bit motes, (2) 32-bit MCUs are a major energy consumer (e.g., tick increases idle current >50 times), comparable to radios, and (3) thread-based concurrency is viable, requiring only 8.3 μs of context switch time. We design a system architecture, based on a tickless multithreading operating system, with cooperative/adaptive clocking, advanced sensor abstraction, and preemptive packet processing. Its efficient MCU control improves concurrency with ~30% less energy consumption. Together, these developments set the system architecture for networked sensors in a new direction.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗