Results for “thread parallelism” — Page 15

DOE OSTI2024

Refining HPCToolkit for application performance analysis at exascale

As part of the US Department of Energy’s Exascale Computing Project (ECP), Rice University has been refining its HPCToolkit performance tools to better support measurement and analysis of applications executing on exascale supercomputers. To efficiently collect performance measurements of GPU-accelerated applications, HPCToolkit employs novel non-blocking data structures to communicate performance measurements between tool threads and application threads. To attribute performance information in detail to source lines, loop nests, and inlined call chains, HPCToolkit performs parallel analysis of large CPU and GPU binaries involved in the execution of an exascale application to rapidly recover mappings between machine instructions and source code. To analyze terabytes of performance measurements gathered during executions at exascale, HPCToolkit employs distributed-memory parallelism, multithreading, sparse data structures, and out-of-core streaming analysis algorithms. To support interactive exploration of profiles up to terabytes in size, HPCToolkit’s hpcviewer graphical user interface uses out-of-core methods to visualize performance data. The result of these efforts is that HPCToolkit now supports collection, analysis, and presentation of profiles and traces of GPU-accelerated applications at exascale. These improvements have enabled HPCToolkit to efficiently measure, analyze and explore terabytes of performance data for executions using as many as 64K MPI ranks and 64K GPU tiles on ORNL’s Frontier supercomputer. HPCToolkit’s support for measurement and analysis of GPU-accelerated applications has been employed to study a collection of open-science applications developed as part of ECP. This paper reports on these experiences, which provided insight into opportunities for tuning applications, strengths and weaknesses of HPCToolkit itself, as well as unexpected behaviors in executions at exascale.

Adhianto, Laksono↗

DOE OSTI2019

Using OpenMP for HEP Framework Algorithm Scheduling [Slides]

The OpenMP standard is the primary mechanism used at high performance computing facilities to allow intra-process parallelization. In contrast, many HEP specific software (such as CMSSW, GaudiHive, and ROOT) make use of Intel's Threading Building Blocks (TBB) library to accomplish the same goal. In this talk we will discuss our work to compare TBB and OpenMP when used for scheduling algorithms to be run by a HEP style data processing framework (i.e. running hundreds of interdependent algorithms at most once for each event read from the detector). This includes both scheduling of different algorithms to be run concurrently as well as scheduling concurrent work within one algorithm. As part of the discussion we present an overview of the OpenMP threading model. We also explain how we used OpenMP when creating a simplified HEP-like processing framework. Using that simplified framework, and a similar one written using TBB, we will present performance comparisons between TBB and different compiler versions of OpenMP.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

NASA NTRS2024

Runtime Thread-Block Optimization for Custom Multistream CUDA Kernels for the Glenn Research Center Communication Analysis Suite

In preparation of the return of humans to the Moon with the coming Artemis missions, NASA scientists must evaluate proposed landing site locations for terrain and communications viability. The Glenn Research Center Communication Analysis Suite (GCAS) combines sophisticated communication network models with accurate lunar terrain to access sites across the Moon’s south pole. Given the importance of proper site selection to crew safety and mission success, many locations need to be analyzed resulting in a large computational load needing to be performed. To meet the growing project demands, development has begun to improve the runtime efficiency of GCAS with GPU parallelization by way of multi-stream CUDA kernels. One of the most prominent factors in kernel optimization is the proper selection of thread-block dimensions in order to maximize the concurrent operation on the device. Typically, thread-block dimensions are optimized by hand requiring many stages of benchmarking and iteration. Additionally, given the main conditions to optimization are the physical GPU architecture and problem size, these optimal dimensions are non-portable and fragile in their scope. As such, a novel optimization routine was developed to generate the optimal thread-block dimensions during runtime with considerations to hardware specifications and problem size resolving the issues of portability and enabling the function of more dynamic routines.

Aden Bergstresser↗

DOE OSTI2020

Using OpenMP for HEP framework algorithm scheduling

The OpenMP standard is the primary mechanism used at high performance computing facilities to allow intra-process parallelization. In contrast, many HEP specific software packages (such as CMSSW, GaudiHive, and ROOT) make use of Intel’s Threading Building Blocks (TBB) library to accomplish the same goal. In these proceedings we will discuss our work to compare TBB and OpenMP when used for scheduling algorithms to be run by a HEP style data processing framework. This includes both scheduling of different interdependent algorithms to be run concurrently as well as scheduling concurrent work within one algorithm. As part of the discussion we present an overview of the OpenMP threading model. We also explain how we used OpenMP when creating a simplified HEP-like processing framework. Using that simplified framework, and a similar one written using TBB, we will present performance comparisons between TBB and different compiler versions of OpenMP.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Fencing data transfers in a parallel active messaging interface of a parallel computer

Fencing data transfers in a parallel active messaging interface (`PAMI`) of a parallel computer, the PAMI including 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 and through data communications resources including at least one segment of shared random access memory; including initiating execution through the PAMI of an ordered sequence of active SEND instructions for SEND data transfers between two endpoints, effecting deterministic SEND data transfers through a segment of shared memory; and executing through the PAMI, with no FENCE accounting for SEND data transfers, an active FENCE instruction, the FENCE instruction completing execution only after completion of all SEND instructions initiated prior to execution of the FENCE instruction for SEND data transfers between the two endpoints.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Fencing data transfers in a parallel active messaging interface of a parallel computer

Fencing data transfers in a parallel active messaging interface (`PAMI`) of a parallel computer, the PAMI including 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 and through data communications resources including at least one segment of shared random access memory; including initiating execution through the PAMI of an ordered sequence of active SEND instructions for SEND data transfers between two endpoints, effecting deterministic SEND data transfers through a segment of shared memory; and executing through the PAMI, with no FENCE accounting for SEND data transfers, an active FENCE instruction, the FENCE instruction completing execution only after completion of all SEND instructions initiated prior to execution of the FENCE instruction for SEND data transfers between the two endpoints.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Fencing data transfers in a parallel active messaging interface of a parallel computer

Fencing data transfers in a parallel active messaging interface (`PAMI`) of a parallel computer, the PAMI including data communications endpoints, each endpoint comprising 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 and through data communications resources including a deterministic data communications network, including initiating execution through the PAMI of an ordered sequence of active SEND instructions for SEND data transfers between two endpoints, effecting deterministic SEND data transfers; and executing through the PAMI, with no FENCE accounting for SEND data transfers, an active FENCE instruction, the FENCE instruction completing execution only after completion of all SEND instructions initiated prior to execution of the FENCE instruction for SEND data transfers between the two endpoints.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Fencing data transfers in a parallel active messaging interface of a parallel computer

Fencing data transfers in a parallel active messaging interface (`PAMI`) of a parallel computer, the PAMI including data communications endpoints, each endpoint comprising 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 and through data communications resources including a deterministic data communications network, including initiating execution through the PAMI of an ordered sequence of active SEND instructions for SEND data transfers between two endpoints, effecting deterministic SEND data transfers; and executing through the PAMI, with no FENCE accounting for SEND data transfers, an active FENCE instruction, the FENCE instruction completing execution only after completion of all SEND instructions initiated prior to execution of the FENCE instruction for SEND data transfers between the two endpoints.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

The Metropolis Monte Carlo method with CUDA enabled Graphic Processing Units

We present a CPU–GPU system for runtime acceleration of large molecular simulations using GPU computation and memory swaps. The memory architecture of the GPU can be used both as container for simulation data stored on the graphics card and as floating-point code target, providing an effective means for the manipulation of atomistic or molecular data on the GPU. To fully take advantage of this mechanism, efficient GPU realizations of algorithms used to perform atomistic and molecular simulations are essential. Our system implements a versatile molecular engine, including inter-molecule interactions and orientational variables for performing the Metropolis Monte Carlo (MMC) algorithm, which is one type of Markov chain Monte Carlo. By combining memory objects with floating-point code fragments we have implemented an MMC parallel engine that entirely avoids the communication time of molecular data at runtime. Our runtime acceleration system is a forerunner of a new class of CPU–GPU algorithms exploiting memory concepts combined with threading for avoiding bus bandwidth and communication. The testbed molecular system used here is a condensed phase system of oligopyrrole chains. A benchmark shows a size scaling speedup of 60 for systems with 210,000 pyrrole monomers. Our implementation can easily be combined with MPI to connect in parallel several CPU–GPU duets. -- Highlights: •We parallelize the Metropolis Monte Carlo (MMC) algorithm on one CPU—GPU duet. •The Adaptive Tempering Monte Carlo employs MMC and profits from this CPU—GPU implementation. •Our benchmark shows a size scaling-up speedup of 62 for systems with 225,000 particles. •The testbed involves a polymeric system of oligopyrroles in the condensed phase. •The CPU—GPU parallelization includes dipole—dipole and Mie—Jones classic potentials.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Fault Diagnosis of Hybrid Computing Systems Using Chaotic-Map Method

Computing systems are becoming increasingly complex with nodes consisting of a combination of multi-core central processing units (CPUs), many integrated core (MIC) and graphics processing unit (GPU) accelerators. These computing units and their interconnections are subject to different classes of hardware and software faults, which should be detected to support mitigation measures. We present the chaotic-map method that uses the exponential divergence and wide Fourier properties of the trajectories, combined with memory allocations and assignments to diagnose component-level faults in these hybrid computing systems. We propose lightweight codes that utilize highly parallel chaotic-map computations tailored to isolate faults in arithmetic units, memory elements and interconnects. The diagnosis module on a node utilizes pthreads to place chaotic-map threads on CPU and MIC cores, and CUDA C and OpenCL kernels on GPU blocks. We present experimental diagnosis results on five multi-core CPUs; one MIC; and, seven GPUs with typical diagnosis run-times under a minute.

Rao, Nageswara↗

DOE OSTI2019

Approximation algorithms in combinatorial scientific computing

We survey recent work on approximation algorithms for computing degree-constrained subgraphs in graphs and their applications in combinatorial scientific computing. The problems we consider include maximization versions of cardinality matching, edge-weighted matching, vertex-weighted matching and edge-weighted $b$ -matching, and minimization versions of weighted edge cover and $b$ -edge cover. Exact algorithms for these problems are impractical for massive graphs with several millions of edges. For each problem we discuss theoretical foundations, the design of several linear or near-linear time approximation algorithms, their implementations on serial and parallel computers, and applications. Our focus is on practical algorithms that yield good performance on modern computer architectures with multiple threads and interconnected processors. We also include information about the software available for these problems.

Mathematics↗

NASA NTRS1999

Parallelization of a Multigrid Incompressible Viscous Cavity Flow Solver Using OpenMP

We describe a multigrid scheme for solving the viscous incompressible driven cavity problem that has been parallelized using OpenMP. The incremental parallelization allowed by OpenMP was of great help during the parallelization process. Results show good parallel efficiencies for reasonable problem sizes on an SGI Origin 2000. Since OpenMP allowed us to specify the number of threads (and in turn processors) at runtime, we were able to improve performance when solving on smaller/coarser meshes. This was accomplished by giving each processor a more reasonable amount of work rather than having many processors work on very small segments of the data (and thereby adding significant overhead).

Roe, Kevin↗

DOE OSTI2014

Data communications in a parallel active messaging interface of a parallel computer

Data communications in a parallel active messaging interface ('PAMI') or a parallel computer, the parallel computer including a plurality of compute nodes that execute a parallel application, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution of a compute node, including specification of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications instruction, the instruction characterized by instruction type, the instruction specifying a transmission of transfer data from the origin endpoint to a target endpoint and transmitting, in accordance witht the instruction type, the transfer data from the origin endpoin to the target endpoint.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Data communications in a parallel active messaging interface of a parallel computer

Data communications 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, endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint; transmitting from the origin endpoint to the first target endpoint a Request-To-Send (`RTS`) message advising the first target endpoint of the location and size of the transfer data; assigning by the first target endpoint to each of a plurality of target endpoints separate portions of the transfer data; and receiving by the plurality of target endpoints the transfer data.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Data communications in a parallel active messaging interface of a parallel computer

Data communications 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, endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint; transmitting from the origin endpoint to the first target endpoint a Request-To-Send (`RTS`) message advising the first target endpoint of the location and size of the transfer data; assigning by the first target endpoint to each of a plurality of target endpoints separate portions of the transfer data; and receiving by the plurality of target endpoints the transfer data.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1999

Aerodynamic Shape Optimization Using A Combined Distributed/Shared Memory Paradigm

Current parallel computational approaches involve distributed and shared memory paradigms. In the distributed memory paradigm, each processor has its own independent memory. Message passing typically uses a function library such as MPI or PVM. In the shared memory paradigm, such as that used on the SGI Origin 2000 machine, compiler directives are used to instruct the compiler to schedule multiple threads to perform calculations. In this paradigm, it must be assured that processors (threads) do not simultaneously access regions of memory in such away that errors would occur. This paper utilizes the latest version of the SGI MPI function library to combine the two parallelization paradigms to perform aerodynamic shape optimization of a generic wing/body.

Cheung, Samson↗

DOE OSTI2017

Parallel Asynchronous Distributed-Memory Maximal Independent Set Algorithm with Work Ordering

The maximal independent set (MIS) graph problem arises in many applications such as computer vision, information theory, molecular biology, and process scheduling. The growing scale of graph data suggests the use of distributed memory hardware as a cost-effective approach to providing necessary compute and memory resources. Existing distributed memory parallel MIS algorithms rely on synchronous communication and use techniques such as subgraph computations. In this paper, we present an asynchronous distributed-memory parallel graph algorithm that relies on a virtual directed acyclic graph (DAG) that is created during the algorithm execution. We introduce two additional algorithms that save computations by ordering generated work. The first algorithm applies ordering globally to reduce computations, and the second algorithm applies ordering locally at the level of threads to minimize the synchronization overhead. We use two different implementations of Luby’s algorithm variants as baseline to compare the performance of the presented algorithms: (1) vertex-centric Luby A and Luby B implementations, and (2) the CombBLAS linear-algebra Luby A implementation. Results show that proposed algorithms outperform both implementations of Luby algorithms, especially in distributed execution. Furthermore, we show that for low-diameter graphs the algorithm that applies global ordering scales better than other algorithms and for high diameter graphs the original asynchronous algorithm and thread-level ordering algorithm show better performance.

maximal independent set (MIS), distributed memory ↗

DOE OSTI2016

Revised Extended Grid Library

The Revised Eolus Grid Library (REGL) is a mesh-tracking library that was developed for use with the MCNP6TM computer code so that (radiation) particles can track on an unstructured mesh. The unstructured mesh is a finite element representation of any geometric solid model created with a state-of-the-art CAE/CAD tool. The mesh-tracking library is written using modern Fortran and programming standards; the library is Fortran 2003 compliant. The library was created with a defined application programmer interface (API) so that it could easily integrate with other particle tracking/transport codes. The library does not handle parallel processing via the message passing interface (mpi), but has been used successfully where the host code handles the mpi calls. The library is thread-safe and supports the OpenMP paradigm. As a library, all features are available through the API and overall a tight coupling between it and the host code is required. Features of the library are summarized with the following list: Can accommodate first and second order 4, 5, and 6-sided polyhedra; any combination of element types may appear in a single geometry model; parts may not contain tetrahedra mixed with other element types; pentahedra and hexahedra can be together in the same part; robust handling of overlaps and gaps; tracks element-to-element to produce path length results at the element level; finds element numbers for a given mesh location; finds intersection points on element faces for the particle tracks; produce a data file for post processing results analysis; reads Abaqus .inp input (ASCII) files to obtain information for the global mesh-model; supports parallel input processing via mpi; and support parallel particle transport by both mpi and OpenMP.

Martz, Roger L.↗