Search NASA⌕ Search

SEARCH · Search NASA

Results for “thread parallelism”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRS2012

High-Performance, Multi-Node File Copies and Checksums for Clustered File Systems

Modern parallel file systems achieve high performance using a variety of techniques, such as striping files across multiple disks to increase aggregate I/O bandwidth and spreading disks across multiple servers to increase aggregate interconnect bandwidth. To achieve peak performance from such systems, it is typically necessary to utilize multiple concurrent readers/writers from multiple systems to overcome various singlesystem limitations, such as number of processors and network bandwidth. The standard cp and md5sum tools of GNU coreutils found on every modern Unix/Linux system, however, utilize a single execution thread on a single CPU core of a single system, and hence cannot take full advantage of the increased performance of clustered file systems. Mcp and msum are drop-in replacements for the standard cp and md5sum programs that utilize multiple types of parallelism and other optimizations to achieve maximum copy and checksum performance on clustered file systems. Multi-threading is used to ensure that nodes are kept as busy as possible. Read/write parallelism allows individual operations of a single copy to be overlapped using asynchronous I/O. Multinode cooperation allows different nodes to take part in the same copy/checksum. Split-file processing allows multiple threads to operate concurrently on the same file. Finally, hash trees allow inherently serial checksums to be performed in parallel. Mcp and msum provide significant performance improvements over standard cp and md5sum using multiple types of parallelism and other optimizations. The total speed-ups from all improvements are significant. Mcp improves cp performance over 27x, msum improves md5sum performance almost 19x, and the combination of mcp and msum improves verified copies via cp and md5sum by almost 22x. These improvements come in the form of drop-in replacements for cp and md5sum, so are easily used and are available for download as open source software at http://mutil.sourceforge.net.

Kolano, Paul Z.↗
DOE OSTI2017

High Performance Computing Based Parallel HIearchical Modal Association Clustering (HPAR HMAC)

For many applications, clustering is a crucial step in order to gain insight into the makeup of a dataset. The best approach to a given problem often depends on a variety of factors, such as the size of the dataset, time restrictions, and soft clustering requirements. The HMAC algorithm seeks to combine the strengths of 2 particular clustering approaches: model-based and linkage-based clustering. One particular weakness of HMAC is its computational complexity. HMAC is not practical for mega-scale data clustering. For high-definition imagery, a user would have to wait months or years for a result; for a 16-megapixel image, the estimated runtime skyrockets to over a decade! To improve the execution time of HMAC, it is reasonable to consider an multi-core implementation that utilizes available system resources. An existing imple-mentation (Ray and Cheng 2014) divides the dataset into N partitions - one for each thread prior to executing the HMAC algorithm. This implementation benefits from 2 types of optimization: parallelization and divide-and-conquer. By running each partition in parallel, the program is able to accelerate computation by utilizing more system resources. Although the parallel implementation provides considerable improvement over the serial HMAC, it still suffers from poor computational complexity, O(N2). Once the maximum number of cores on a system is exhausted, the program exhibits slower behavior. We now consider a modification to HMAC that involves a recursive partitioning scheme. Our modification aims to exploit divide-and-conquer benefits seen by the parallel HMAC implementation. At each level in the recursion tree, partitions are divided into 2 sub-partitions until a threshold size is reached. When the partition can no longer be divided without falling below threshold size, the base HMAC algorithm is applied. This results in a significant speedup over the parallel HMAC.

Patlolla, Dilip R↗
NASA NTRS1990

Threaded average temperature thermocouple

A threaded average temperature thermocouple 11 is provided to measure the average temperature of a test situs of a test material 30. A ceramic insulator rod 15 with two parallel holes 17 and 18 through the length thereof is securely fitted in a cylinder 16, which is bored along the longitudinal axis of symmetry of threaded bolt 12. Threaded bolt 12 is composed of material having thermal properties similar to those of test material 30. Leads of a thermocouple wire 20 leading from a remotely situated temperature sensing device 35 are each fed through one of the holes 17 or 18, secured at head end 13 of ceramic insulator rod 15, and exit at tip end 14. Each lead of thermocouple wire 20 is bent into and secured in an opposite radial groove 25 in tip end 14 of threaded bolt 12. Resulting threaded average temperature thermocouple 11 is ready to be inserted into cylindrical receptacle 32. The tip end 14 of the threaded average temperature thermocouple 11 is in intimate contact with receptacle 32. A jam nut 36 secures the threaded average temperature thermocouple 11 to test material 30.

Ward, Stanley W.↗
DOE OSTI2014

Endpoint-based parallel data processing in a parallel active messaging interface of a parallel computer

Endpoint-based parallel data processing in a parallel active messaging interface (`PAMI`) of a parallel computer, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes coupled for data communications through the PAMI, including establishing a data communications geometry, the geometry specifying, for tasks representing processes of execution of the parallel application, a set of endpoints that are used in collective operations of the PAMI including a plurality of endpoints for one of the tasks; receiving in endpoints of the geometry an instruction for a collective operation; and executing the instruction for a collective operation through the endpoints in dependence upon the geometry, including dividing data communications operations among the plurality of endpoints for one of the tasks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

CMS readiness for multi-core workload scheduling

In the present run of the LHC, CMS data reconstruction and simulation algorithms benefit greatly from being executed as multiple threads running on several processor cores. The complexity of the Run 2 events requires parallelization of the code to reduce the memory-per- core footprint constraining serial execution programs, thus optimizing the exploitation of present multi-core processor architectures. The allocation of computing resources for multi-core tasks, however, becomes a complex problem in itself. The CMS workload submission infrastructure employs multi-slot partitionable pilots, built on HTCondor and GlideinWMS native features, to enable scheduling of single and multi-core jobs simultaneously. This provides a solution for the scheduling problem in a uniform way across grid sites running a diversity of gateways to compute resources and batch system technologies. This paper presents this strategy and the tools on which it has been implemented. The experience of managing multi-core resources at the Tier-0 and Tier-1 sites during 2015, along with the deployment phase to Tier-2 sites during early 2016 is reported. The process of performance monitoring and optimization to achieve efficient and flexible use of the resources is also described.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

A Work-Efficient Parallel Sparse Matrix-Sparse Vector Multiplication Algorithm

We design and develop a work-efficient multithreaded algorithm for sparse matrix-sparse vector multiplication (SpMSpV) where the matrix, the input vector, and the output vector are all sparse. SpMSpV is an important primitive in the emerging GraphBLAS standard and is the workhorse of many graph algorithms including breadth-first search, bipartite graph matching, and maximal independent set. As thread counts increase, existing multithreaded SpMSpV algorithms can spend more time accessing the sparse matrix data structure than doing arithmetic. Our shared-memory parallel SpMSpV algorithm is work efficient in the sense that its total work is proportional to the number of arithmetic operations required. The key insight is to avoid each thread individually scan the list of matrix columns. Our algorithm is simple to implement and operates on existing column-based sparse matrix formats. It performs well on diverse matrices and vectors with heterogeneous sparsity patterns. A high-performance implementation of the algorithm attains up to 15x speedup on a 24-core Intel Ivy Bridge processor and up to 49x speedup on a 64-core Intel KNL manycore processor. In contrast to implementations of existing algorithms, the performance of our algorithm is sustained on a variety of different input types include matrices representing scale-free and high-diameter graphs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Community Detection on the GPU

We present and evaluate a new GPU algorithm based on the Louvain method for community detection. Our algorithm is the first for this problem that parallelizes the access to individual edges. In this way we can fine tune the load balance when processing networks with nodes of highly varying degrees. This is achieved by scaling the number of threads assigned to each node according to its degree. Extensive experiments show that we obtain speedups up to a factor of 270 compared to the sequential algorithm. The algorithm consistently outperforms other recent shared memory implementations and is only one order of magnitude slower than the current fastest parallel Louvain method running on a Blue Gene/Q supercomputer using more than 500K threads.

community detection↗
DOE OSTI2021

Theoretically and practically efficient parallel nucleus decomposition

This paper studies the nucleus decomposition problem, which has been shown to be useful in finding dense substructures in graphs. We present a novel parallel algorithm that is efficient both in theory and in practice. Our algorithm achieves a work complexity matching the best sequential algorithm while also having low depth (parallel running time), which significantly improves upon the only existing parallel nucleus decomposition algorithm (Sariyüce et al. , PVLDB 2018). The key to the theoretical efficiency of our algorithm is a new lemma that bounds the amount of work done when peeling cliques from the graph, combined with the use of a theoretically-efficient parallel algorithms for clique listing and bucketing. We introduce several new practical optimizations, including a new multi-level hash table structure to store information on cliques space-efficiently and a technique for traversing this structure cache-efficiently. On a 30-core machine with two-way hyper-threading on real-world graphs, we achieve up to a 55x speedup over the state-of-the-art parallel nucleus decomposition algorithm by Sariyüce et al. , and up to a 40x self-relative parallel speedup. We are able to efficiently compute larger nucleus decompositions than prior work on several million-scale graphs for the first time.

Computer Science↗
DOE OSTI2020

Fast shared-memory streaming multilevel graph partitioning

In this report we show that a fast parallel graph partitioner can benefit many applications by reducing data transfers. The online methods for partitioning graphs have to be fast and they often rely on simple one-pass streaming algorithms, while the offline methods for partitioning graphs contain more involved algorithms and the most successful methods in this category belong to the multilevel approaches. In this work, we assess the feasibility of using streaming graph partitioning algorithms within the multilevel framework. Our end goal is to come up with a fast parallel offline multilevel partitioner that can produce competitive cutsize quality. We rely on a simple but fast and flexible streaming algorithm throughout the entire multilevel framework. This streaming algorithm serves multiple purposes in the partitioning process: a clustering algorithm in the coarsening, an effective algorithm for the initial partitioning, and a fast refinement algorithm in the uncoarsening. Its simple nature also lends itself easily for parallelization. The experiments on various graphs show that our approach is on the average up to 5.1x faster than the multi-threaded MeTiS, which comes at the expense of only 2x worse cutsize.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

LAMMPS strong scaling performance optimization on Blue Gene/Q

LAMMPS "Large-scale Atomic/Molecular Massively Parallel Simulator" is an open-source molecular dynamics package from Sandia National Laboratories. Significant performance improvements in strong-scaling and time-to-solution for this application on IBM's Blue Gene/Q have been achieved through computational optimizations of the OpenMP versions of the short-range Lennard-Jones term of the CHARMM force field and the long-range Coulombic interaction implemented with the PPPM (particle-particle-particle mesh) algorithm, enhanced by runtime parameter settings controlling thread utilization. Additionally, MPI communication performance improvements were made to the PPPM calculation by re-engineering the parallel 3D FFT to use MPICH collectives instead of point-to-point. Performance testing was done using an 8.4-million atom simulation scaling up to 16 racks on the Mira system at Argonne Leadership Computing Facility (ALCF). Speedups resulting from this effort were in some cases over 2x.

Coffman, Paul↗
NASA NTRS2000

Testing New Programming Paradigms with NAS Parallel Benchmarks

Over the past decade, high performance computing has evolved rapidly, not only in hardware architectures but also with increasing complexity of real applications. Technologies have been developing to aim at scaling up to thousands of processors on both distributed and shared memory systems. Development of parallel programs on these computers is always a challenging task. Today, writing parallel programs with message passing (e.g. MPI) is the most popular way of achieving scalability and high performance. However, writing message passing programs is difficult and error prone. Recent years new effort has been made in defining new parallel programming paradigms. The best examples are: HPF (based on data parallelism) and OpenMP (based on shared memory parallelism). Both provide simple and clear extensions to sequential programs, thus greatly simplify the tedious tasks encountered in writing message passing programs. HPF is independent of memory hierarchy, however, due to the immaturity of compiler technology its performance is still questionable. Although use of parallel compiler directives is not new, OpenMP offers a portable solution in the shared-memory domain. Another important development involves the tremendous progress in the internet and its associated technology. Although still in its infancy, Java promisses portability in a heterogeneous environment and offers possibility to "compile once and run anywhere." In light of testing these new technologies, we implemented new parallel versions of the NAS Parallel Benchmarks (NPBs) with HPF and OpenMP directives, and extended the work with Java and Java-threads. The purpose of this study is to examine the effectiveness of alternative programming paradigms. NPBs consist of five kernels and three simulated applications that mimic the computation and data movement of large scale computational fluid dynamics (CFD) applications. We started with the serial version included in NPB2.3. Optimization of memory and cache usage was applied to several benchmarks, noticeably BT and SP, resulting in better sequential performance. In order to overcome the lack of an HPF performance model and guide the development of the HPF codes, we employed an empirical performance model for several primitives found in the benchmarks. We encountered a few limitations of HPF, such as lack of supporting the "REDISTRIBUTION" directive and no easy way to handle irregular computation. The parallelization with OpenMP directives was done at the outer-most loop level to achieve the largest granularity. The performance of six HPF and OpenMP benchmarks is compared with their MPI counterparts for the Class-A problem size in the figure in next page. These results were obtained on an SGI Origin2000 (195MHz) with MIPSpro-f77 compiler 7.2.1 for OpenMP and MPI codes and PGI pghpf-2.4.3 compiler with MPI interface for HPF programs.

Jin, H.↗
DOE OSTI2016

Scalable Metropolis Monte Carlo for simulation of hard shapes

We design and implement a scalable hard particle Monte Carlo simulation toolkit (HPMC), and release it open source as part of HOOMD-blue. HPMC runs in parallel on many CPUs and many GPUs using domain decomposition. We employ BVH trees instead of cell lists on the CPU for fast performance, especially with large particle size disparity, and optimize inner loops with SIMD vector intrinsics on the CPU. Our GPU kernel proposes many trial moves in parallel on a checkerboard and uses a block-level queue to redistribute work among threads and avoid divergence. HPMC supports a wide variety of shape classes, including spheres/disks, unions of spheres, convex polygons, convex spheropolygons, concave polygons, ellipsoids/ellipses, convex polyhedra, convex spheropolyhedra, spheres cut by planes, and concave polyhedra. NVT and NPT ensembles can be run in 2D or 3D triclinic boxes. Additional integration schemes permit Frenkel–Ladd free energy computations and implicit depletant simulations. In a benchmark system of a fluid of 4096 pentagons, HPMC performs 10 million sweeps in 10 min on 96 CPU cores on XSEDE Comet. The same simulation would take 7.6 h in serial. HPMC also scales to large system sizes, and the same benchmark with 16.8 million particles runs in 1.4 h on 2048 GPUs on OLCF Titan.

Computer Science↗
DOE OSTI2023

Code modernization strategies for short-range non-bonded molecular dynamics simulations

Modern HPC systems are increasingly relying on greater core counts and wider vector registers. Thus, applications need to be adapted to fully utilize these hardware capabilities. One class of applications that can benefit from this increase in parallelism are molecular dynamics simulations. In this paper, we describe our efforts at modernizing the ESPResSo++ simulation package for molecular dynamics by restructuring its particle data layout for efficient memory accesses and applying vectorization techniques to benefit the calculation of short-range non-bonded forces, which results in an overall three times speedup and serves as a baseline for further optimizations. We also implement fine-grained parallelism for multi-core CPUs through HPX, a C++ runtime system which uses lightweight threads and an asynchronous many-task approach to maximize concurrency. Our goal is to evaluate the performance of an HPX-based approach compared to the bulk-synchronous MPI-based implementation. This requires the introduction of an additional layer to the domain decomposition scheme that defines the task granularity. On spatially inhomogeneous systems, which impose a corresponding load-imbalance in traditional MPI-based approaches, we demonstrate that by choosing an optimal task size, the efficient work-stealing mechanisms of HPX can overcome the overhead of communication resulting in an overall 1.4 times speedup compared to the baseline MPI version.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

UPC++ Programmer’s Guide (v1.0 2017.9)

UPC++ is a C++11 library that provides Asynchronous Partitioned Global Address Space (APGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The APGAS model is single program, multiple-data (SPMD), with each separate thread of execution (referred to as a rank, a term borrowed from MPI) having access to local memory as it would in C++. However, APGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the ranks. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

UPC++ Programmer’s Guide, v1.0-2018.3.0

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate thread of execution (referred to as a rank, a term borrowed from MPI) having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the ranks. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Optimizing the hit finding algorithm for liquid argon TPC neutrino detectors using parallel architectures

Neutrinos are particles that interact rarely, so identifying them requires large detectors which produce lots of data. Processing this data with the computing power available is becoming even more difficult as the detectors increase in size to reach their physics goals. Liquid argon time projection chamber (LArTPC) neutrino experiments are expected to grow in the next decade to have 100 times more wires than in currently operating experiments, and modernization of LArTPC reconstruction code, including parallelization both at data- and instruction-level, will help to mitigate this challenge. The LArTPC hit finding algorithm is used across multiple experiments through a common software framework. In this paper we discuss a parallel implementation of this algorithm. Using a standalone setup we find speedup factors of two times from vectorization and 30–100 times from multi-threading on Intel architectures. The new version has been incorporated back into the framework so that it can be used by experiments. On a serial execution, the integrated version is about 10 times faster than the previous one and, once parallelization is enabled, more speedups comparable to the standalone program are achieved.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2016

Asynchronous In Situ Connected-Components Analysis for Complex Fluid flows, In: ISAV '16 Proceedings of the 2nd Workshop on In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization

The simulation of multiscale physics is an important challenge for scientific computing. For this class of problem, large three-dimensional simulations are performed to advance scientific inquiry. On massively parallel computing systems, the volume of data generated by such approaches can become a productivity bottleneck if the raw data generated from the simulation is analyzed in a post-processing step. To address this, we present a physics-based framework for in situ data reduction that is theoretically grounded in multiscale averaging theory. We show how task parallelism can be exploited to concurrently perform a variety of analysis tasks with data-dependent costs, including the generation of iso-surfaces, morphological analyses, and connected components analysis. All analyses are performed in parallel using distributed memory and use the same domain decomposition as the simulation. A task management framework is constructed to leverage available parallelism within a node for analysis. The capabilities of the framework are to launch asynchronous analysis threads, manage dependencies between different tasks, promote data locality and minimize the impact of data transfers. The framework is applied to analyze GPU-based simulations of two-fluid-phase flow in porous media, generating a set of averaged measures that represents the overall system behavior. We demonstrate how the approach can be applied to perform physically-consistent analysis over fluid sub-regions determined from connected components analysis. Simulations performed on Oak Ridge National Lab's Titan supercomputer are profiled to demonstrate the performance of the associated multi-threaded in situ analysis approach for typical production simulation of two-fluid-phase flow.

Computer Science↗
DOE OSTI2015

The SprayList: a scalable relaxed priority queue

High-performance concurrent priority queues are essential for applications such as task scheduling and discrete event simulation. Unfortunately, even the best performing implementations do not scale past a number of threads in the single digits. This is because of the sequential bottleneck in accessing the elements at the head of the queue in order to perform a DeleteMin operation. In this paper, we present the SprayList, a scalable priority queue with relaxed ordering semantics. Starting from a non-blocking SkipList, the main innovation behind our design is that the DeleteMin operations avoid a sequential bottleneck by ``spraying'' themselves onto the head of the SkipList list in a coordinated fashion. The spraying is implemented using a carefully designed random walk, so that DeleteMin returns an element among the first O(p log^3 p) in the list, with high probability, where p is the number of threads. We prove that the running time of a DeleteMin operation is O(log^3 p), with high probability, independent of the size of the list. Our experiments show that the relaxed semantics allow the data structure to scale for high thread counts, comparable to a classic unordered SkipList. As a result, we observe that, for reasonably parallel workloads, the scalability benefits of relaxation considerably outweigh the additional work due to out-of-order execution.

97 MATHEMATICS AND COMPUTING↗