Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel computer architecture”

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 757 records · Page 42

Scalable Triadic Analysis of Large-Scale Graphs: Multi-Core vs. Multi-Processor vs. Multi-Threaded Shared Memory Architectures

Triadic analysis encompasses a useful set of graph mining methods that is centered on the concept of a triad, which is a subgraph of three nodes and the configuration of directed edges across the nodes. Such methods are often applied in the social sciences as well as many other diverse fields. Triadic methods commonly operate on a triad census that counts the number of triads of every possible edge configuration in a graph. Like other graph algorithms, triadic census algorithms do not scale well when graphs reach tens of millions to billions of nodes. To enable the triadic analysis of large-scale graphs, we developed and optimized a triad census algorithm to efficiently execute on shared memory architectures. We will retrace the development and evolution of a parallel triad census algorithm. Over the course of several versions, we continually adapted the code’s data structures and program logic to expose more opportunities to exploit parallelism on shared memory that would translate into improved computational performance. We will recall the critical steps and modifications that occurred during code development and optimization. Furthermore, we will compare the performances of triad census algorithm versions on three specific systems: Cray XMT, HP Superdome, and AMD multi-core NUMA machine. These three systems have shared memory architectures but with markedly different hardware capabilities to manage parallelism.

Triadic analysis, triads, triad census, large-scal↗

Vectorization of a particle simulation method for hypersonic rarefied flow

An efficient particle simulation technique for hypersonic rarefied flows is presented at an algorithmic and implementation level. The implementation is for a vector computer architecture, specifically the Cray-2. The method models an ideal diatomic Maxwell molecule with three translational and two rotational degrees of freedom. Algorithms are designed specifically for compatibility with fine grain parallelism by reducing the number of data dependencies in the computation. By insisting on this compatibility, the method is capable of performing simulation on a much larger scale than previously possible. A two-dimensional simulation of supersonic flow over a wedge is carried out for the near-continuum limit where the gas is in equilibrium and the ideal solution can be used as a check on the accuracy of the gas model employed in the method. Also, a three-dimensional, Mach 8, rarefied flow about a finite-span flat plate at a 45 degree angle of attack was simulated. It utilized over 10 to the 7th particles carried through 400 discrete time steps in less than one hour of Cray-2 CPU time. This problem was chosen to exhibit the capability of the method in handling a large number of particles and a true three-dimensional geometry.

Mcdonald, Jeffrey D.↗

A biconjugate gradient type algorithm on massively parallel architectures

The biconjugate gradient (BCG) method is the natural generalization of the classical conjugate gradient algorithm for Hermitian positive definite matrices to general non-Hermitian linear systems. Unfortunately, the original BCG algorithm is susceptible to possible breakdowns and numerical instabilities. Recently, Freund and Nachtigal have proposed a novel BCG type approach, the quasi-minimal residual method (QMR), which overcomes the problems of BCG. Here, an implementation is presented of QMR based on an s-step version of the nonsymmetric look-ahead Lanczos algorithm. The main feature of the s-step Lanczos algorithm is that, in general, all inner products, except for one, can be computed in parallel at the end of each block; this is unlike the other standard Lanczos process where inner products are generated sequentially. The resulting implementation of QMR is particularly attractive on massively parallel SIMD architectures, such as the Connection Machine.

Freund, Roland W.↗

A feasibility study on porting the community land model onto accelerators using OpenACC

As environmental models (such as Accelerated Climate Model for Energy (ACME), Parallel Reactive Flow and Transport Model (PFLOTRAN), Arctic Terrestrial Simulator (ATS), etc.) became more and more complicated, we are facing enormous challenges regarding to porting those applications onto hybrid computing architecture. OpenACC appears as a very promising technology, therefore, we have conducted a feasibility analysis on porting the Community Land Model (CLM), a terrestrial ecosystem model within the Community Earth System Models (CESM)). Specifically, we used automatic function testing platform to extract a small computing kernel out of CLM, then we apply this kernel into the actually CLM dataflow procedure, and investigate the strategy of data parallelization and the benefit of data movement provided by current implementation of OpenACC. Even it is a non-intensive kernel, on a single 16-core computing node, the performance (based on the actual computation time using one GPU) of OpenACC implementation is 2.3 time faster than that of OpenMP implementation using single OpenMP thread, but it is 2.8 times slower than the performance of OpenMP implementation using 16 threads. On multiple nodes, MPI_OpenACC implementation demonstrated very good scalability on up to 128 GPUs on 128 computing nodes. This study also provides useful information for us to look into the potential benefits of “deep copy” capability and “routine” feature of OpenACC standards. In conclusion, we believe that our experience on the environmental model, CLM, can be beneficial to many other scientific research programs who are interested to porting their large scale scientific code using OpenACC onto high-end computers, empowered by hybrid computing architecture.

54 ENVIRONMENTAL SCIENCES↗

Implementing Molecular Dynamics on Hybrid High Performance Computers - Particle-Particle Particle-Mesh

The use of accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high-performance computers, machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. In this paper, we present a continuation of previous work implementing algorithms for using accelerators into the LAMMPS molecular dynamics software for distributed memory parallel hybrid machines. In our previous work, we focused on acceleration for short-range models with an approach intended to harness the processing power of both the accelerator and (multi-core) CPUs. To augment the existing implementations, we present an efficient implementation of long-range electrostatic force calculation for molecular dynamics. Specifically, we present an implementation of the particle-particle particle-mesh method based on the work by Harvey and De Fabritiis. We present benchmark results on the Keeneland InfiniBand GPU cluster. We provide a performance comparison of the same kernels compiled with both CUDA and OpenCL. We discuss limitations to parallel efficiency and future directions for improving performance on hybrid or heterogeneous computers.

97 MATHEMATICS AND COMPUTING↗

Efficient parallel solution of parabolic equations - Implicit methods on the Cedar multicluster

A class of implicit methods for the parallel solution of linear parabolic differential equations based on Pade and Chebyshev rational approximations to the matrix exponential are presented. It is pointed out that this approach incorporates both natural hierarchical parallelism, improved intrinsic efficiency, and fewer timesteps. These advantages lead to an extremely fast family of methods for the solution of certain time-dependent problems. These techniques are illustrated with numerical experiments on the University of Illinois Cedar multicluster architecture. The experiments indicate that implicit methods of very high degree offer great promise for the solution of certain parabolic problems when in computational environment with parallel resources. Hierarchically organized parallel computers, such as the Cedar multicluster, are found to be especially attractive for these schemes.

Gallopoulos, E.↗

ECP ST Capability Assessment Report (CAR) for VTK-m (FY20)

The ECP/VTK-m project is providing the core capabilities to perform scientific visualization on Exascale architectures. The ECP/VTK-m project fills the critical feature gap of performing visualization and analysis on processors like graphics-based processors. The results of this project will be delivered in tools like ParaView, Vislt, and Ascent as well as in stand-alone form. Moreover, these projects are depending on this ECP effort to be able to make effective use of ECP architectures. One of the biggest recent changes in high-performance computing is the increasing use of accelerators. Accelerators contain processing cores that independently are inferior to a core in a typical CPU, but these cores are replicated and grouped such that their aggregate execution provides a very high computation rate at a much lower power. Current and future CPU processors also require much more explicit parallelism. Each successive version of the hardware packs more cores into each processor, and technologies like hyper threading and vector operations require even more parallel processing to leverage each core's full potential. VTK-m is a toolkit of scientific visualization algorithms for emerging processor architectures. VTK-m supports the fine-grained concurrency for data analysis and visualization algorithms required to drive extreme scale computing by providing abstract models for data and execution that can be applied to a variety of algorithms across many different processor architectures. The ECP/VTK-m project is building up the VTK-m codebase with the necessary visualization algorithm implementations that run across the varied hardware platforms to be leveraged at the Exascale. We will be working with other ECP projects, such as ALPINE, to integrate the new VTK-m code into production software to enable visualization on our HPC systems.

97 MATHEMATICS AND COMPUTING↗

Multi-task Parallelism for Robust Pre-training of Graph Foundation Models on Multi-source, Multi-fidelity Atomistic Modeling Data

Graph foundation models using graph neural networks promise sustainable, efficient atomistic modeling. To tackle challenges of processing multi-source, multi-fidelity data during pre-training, recent studies employ multi-task learning, in which shared message passing layers initially process input atomistic structures regardless of source, then route them to multiple decoding heads that predict data-specific outputs. This approach stabilizes pre-training and enhances a model’s transferability to unexplored chemical regions. Preliminary results on approximately four million structures are encouraging, yet questions remain about generalizability to larger, more diverse datasets and scalability on supercomputers. We propose a multi-task parallelism method that distributes each head across computing resources with GPU acceleration. Implemented in the open-source HydraGNN architecture, our method was trained on over 24 million structures from five datasets and tested on the Perlmutter, Aurora, and Frontier supercomputers, demonstrating efficient scaling on all three highly heterogeneous super-computing architectures.

Lupo Pasini, Massimiliano [ORNL] (ORCID:0000000249↗

Mesoscopic Modeling and Rapid Simulation of Incremental Changes in Epidemic Scenarios on GPUs

In simulation-based studies and analyses of epidemics, a major challenge lies in resolving the conflict between fidelity of models and the speed of their simulation. Another related challenge arises in dealing with the large number of what–if scenarios that need to be explored. Here, we describe new computational methods that together provide an approach to dealing with both challenges. A mesoscopic modeling approach is described that strikes a middle ground between macroscopic models based on coupled differential equations and microscopic models built on fine-grained behaviors at the individual entity level. The mesoscopic approach offers the ability to incorporate complex compositions of multiple layers of dynamics even while retaining the potential for aggregate behaviors at varying levels. It also is an excellent match to the accelerator-based architectures of modern computing platforms in which graphical processing units (GPUs) can be exploited for fast simulation via the parallel execution mode of single instruction multiple thread (SIMT). The challenge of simulating a large number of scenarios is addressed via a method of sharing model state and computation across a tree of what–if scenarios that are localized, incremental changes to a large base simulation. A combination of the mesoscopic modeling approach and the incremental what–if scenario tree evaluation has been implemented in the software on modern GPUs. Synthetic simulation scenarios are presented to demonstrate the computational characteristics of our approach. Results from the experiments with large population data, including USA, UK, and India, illustrate the modeling methodology and computational performance on thousands of synthetically generated what–if scenarios. Execution of our implementation scaled to 8192 GPUs of supercomputing platforms demonstrates the ability to rapidly evaluate what–if scenarios several orders of magnitude faster than the conventional methods.

97 MATHEMATICS AND COMPUTING↗

Designing the Cloud-based DOE Systems Biology Knowledgebase

Systems Biology research, even more than many other scientific domains, is becoming increasingly data-intensive. Not only have advances in experimental and computational technologies lead to an exponential increase in scientific data volumes and their complexity, but increasingly such databases themselves are providing the basis for new scientific discoveries. To engage effectively with these community resources, integrated analyses, synthesis and simulation software is needed, regularly supported by scientific workflows. In order to provide a more collaborative, community driven research environment for this heterogeneous setting, the Department of Energy (DOE) has decided to develop a federated, cloud based cyber infrastructure - the Systems Biology Knowledgebase (Kbase). Pacific Northwest National Laboratory (PNNL) with its long tradition in data intensive science lead two of the five initial pilot projects, these two focusing on defining and testing the basic federated cloud-based system architecture and develop a prototype implementation. Hereby the community wide accessibility of biological data and the capability to integrate and analyze this data within its changing research context were seen as key technical functionalities the Kbase needed to enable. In this paper we describe the results of our investigations into the design of a cloud based federated infrastructure for: (1) Semantics driven data discovery, access and integration; (2) Data annotation, publication and sharing; (3) Workflow enabled data analysis; and (4) Project based collaborative working. We describe our approach, exemplary use cases and our prototype implementation that demonstrates the feasibility of this approach.

99 GENERAL AND MISCELLANEOUS↗

Managing genomic variant calling workflows with Swift/T

Bioinformatics research is frequently performed using complex workflows with multiple steps, fans, merges, and conditionals. This complexity makes management of the workflow difficult on a computer cluster, especially when running in parallel on large batches of data: hundreds or thousands of samples at a time. Scientific workflow management systems could help with that. Many are now being proposed, but is there yet the “best” workflow management system for bioinformatics? Such a system would need to satisfy numerous, sometimes conflicting requirements: from ease of use, to seamless deployment at peta- and exa-scale, and portability to the cloud. We evaluated Swift/T as a candidate for such role by implementing a primary genomic variant calling workflow in the Swift/T language, focusing on workflow management, performance and scalability issues that arise from production-grade big data genomic analyses. In the process we introduced novel features into the language, which are now part of its open repository. Additionally, we formalized a set of design criteria for quality, robust, maintainable workflows that must function at-scale in a production setting, such as a large genomic sequencing facility or a major hospital system. The use of Swift/T conveys two key advantages. (1) It operates transparently in multiple cluster scheduling environments (PBS Torque, SLURM, Cray aprun environment, etc.), thus a single workflow is trivially portable across numerous clusters. (2) The leaf functions of Swift/T permit developers to easily swap executables in and out of the workflow, which makes it easy to maintain and to request resources optimal for each stage of the pipeline. While Swift/T’s data-level parallelism eliminates the need to code parallel analysis of multiple samples, it does make debugging more difficult, as is common for implicitly parallel code. Nonetheless, the language gives users a powerful and portable way to scale up analyses in many computing architectures.

Science & Technology - Other Topics↗

Mapping a battlefield simulation onto message-passing parallel architectures

Perhaps the most critical problem in distributed simulation is that of mapping: without an effective mapping of workload to processors the speedup potential of parallel processing cannot be realized. Mapping a simulation onto a message-passing architecture is especially difficult when the computational workload dynamically changes as a function of time and space; this is exactly the situation faced by battlefield simulations. This paper studies an approach where the simulated battlefield domain is first partitioned into many regions of equal size; typically there are more regions than processors. The regions are then assigned to processors; a processor is responsible for performing all simulation activity associated with the regions. The assignment algorithm is quite simple and attempts to balance load by exploiting locality of workload intensity. The performance of this technique is studied on a simple battlefield simulation implemented on the Flex/32 multiprocessor. Measurements show that the proposed method achieves reasonable processor efficiencies. Furthermore, the method shows promise for use in dynamic remapping of the simulation.

Nicol, David M.↗

Multidisciplinary computational aerosciences

By combining CFD with the computational state-of-the-art in structural mechanics, propulsion, aeroacoustics, electromagnetics, etc., multidisciplinary computational aerosciences (MCAS) poses a formidable computational challenge requiring the use of massively parallel machines. Research efforts must accordingly be directed to three areas: (1) parallel architectures, (2) systems software, and (3) applications software; attention is presently given to the last of these, in view of developments at NASA-Ames in solution methodology, physical modeling, and multidisciplinary validation experiments.

Kutler, P.↗

Task scheduling in dataflow computer architectures

Dataflow computers provide a platform for the solution of a large class of computational problems, which includes digital signal processing and image processing. Many typical applications are represented by a set of tasks which can be repetitively executed in parallel as specified by an associated dataflow graph. Research in this area aims to model these architectures, develop scheduling procedures, and predict the transient and steady state performance. Researchers at NASA have created a model and developed associated software tools which are capable of analyzing a dataflow graph and predicting its runtime performance under various resource and timing constraints. These models and tools were extended and used in this work. Experiments using these tools revealed certain properties of such graphs that require further study. Specifically, the transient behavior at the beginning of the execution of a graph can have a significant effect on the steady state performance. Transformation and retiming of the application algorithm and its initial conditions can produce a different transient behavior and consequently different steady state performance. The effect of such transformations on the resource requirements or under resource constraints requires extensive study. Task scheduling to obtain maximum performance (based on user-defined criteria), or to satisfy a set of resource constraints, can also be significantly affected by a transformation of the application algorithm. Since task scheduling is performed by heuristic algorithms, further research is needed to determine if new scheduling heuristics can be developed that can exploit such transformations. This work has provided the initial development for further long-term research efforts. A simulation tool was completed to provide insight into the transient and steady state execution of a dataflow graph. A set of scheduling algorithms was completed which can operate in conjunction with the modeling and performance tools previously developed. Initial studies on the performance of these algorithms were done to examine the effects of application algorithm transformations as measured by such quantities as number of processors, time between outputs, time between input and output, communication time, and memory size.

Katsinis, Constantine↗

Mapping Dense LU Factorization on Multicore Supercomputer Nodes

Dense LU factorization is a prominent benchmark used to rank the performance of supercomputers. Many implementations use block-cyclic distributions of matrix blocks onto a two-dimensional process grid. The process grid dimensions drive a trade-off between communication and computation and are architecture- and implementation-sensitive. The critical panel factorization steps can be made less communication-bound by overlapping asynchronous collectives for pivoting with the computation of rank-k updates. By shifting the computation-communication trade-off, a modified block-cyclic distribution can beneficially exploit more available parallelism on the critical path, and reduce panel factorization's memory hierarchy contention on now-ubiquitous multicore architectures. During active panel factorization, rank-1 updates stream through memory with minimal reuse. In a column-major process grid, the performance of this access pattern degrades as too many streaming processors contend for access to memory. A block-cyclic mapping in the row-major order does not encounter this problem, but consequently sacrifices node and network locality in the critical pivoting steps. We introduce 'striding' to vary between the two extremes of row- and column-major process grids. The maximum available parallelism in the critical path work (active panel factorization, triangular solves, and subsequent broadcasts) is bounded by the length or width of the process grid. Increasing one dimension of the process grid decreases the number of distinct processes and nodes in the other dimension. To increase the harnessed parallelism in both dimensions, we start with a tall process grid. We then apply periodic 'rotation' to this grid to restore exploited parallelism along the row to previous levels. As a test-bed for further mapping experiments, we describe a dense LU implementation that allows a block distribution to be defined as a general function of block to processor. Other mappings can be tested wit...

Computer Science↗