Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Multicore”

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 271 records · Page 15

IRIS: A Performance-Portable Framework for Cross-Platform Heterogeneous Computing

From edge to exascale, computer architectures are becoming more heterogeneous and complex. The systems typically have fat nodes, with multicore CPUs and multiple hardware accelerators such as GPUs, FPGAs, and DSPs. This complexity is causing a crisis in programming systems and performance portability. Several programming systems are working to address these challenges, but the increasing architectural diversity is forcing software stacks and applications to be specialized for each architecture. As we show, all of these approaches critically depend on their software framework for discovery, execution, scheduling, and data orchestration. To address this challenge, we believe that a more agile and proactive software framework is essential to increase performance portability and improve user productivity. To this end, we have designed and implemented IRIS: a performance-portable framework for cross-platform heterogeneous computing. IRIS can discover available resources, manage multiple diverse programming platforms (e.g., CUDA, Hexagon, HIP, Level Zero, OpenCL, OpenMP) simultaneously in the same execution, respect data dependencies, orchestrate data movement proactively, and provide for user-configurable scheduling. To simplify data movement, IRIS introduces a shared virtual device memory with relaxed consistency among different heterogeneous devices. IRIS also adds an automatic kernel workload partitioning technique using the polyhedral model so that it can resize kernels for a wide range of devices. Our evaluation on three architectures, ranging from Qualcomm Snapdragon to a Summit supercomputer node, shows that IRIS improves portability across a wide range of diverse heterogeneous architectures with negligible overhead.

97 MATHEMATICS AND COMPUTING↗

Grouper: A Compact, Streamable Triangle Mesh Data Structure

Here, we present Grouper: an all-in-one compact file format, random-access data structure, and streamable representation for large triangle meshes. Similarly to the recently published SQuad representation, Grouper represents the geometry and connectivity of a mesh by grouping vertices and triangles into fixed-size records, most of which store two adjacent triangles and a shared vertex. Unlike SQuad, however, Grouper interleaves geometry with connectivity and uses a new connectivity representation to ensure that vertices and triangles can be stored in a coherent order that enables memory-efficient sequential stream processing. We also present a linear-time construction algorithm that allows streaming out Grouper meshes using a small memory footprint while preserving the initial ordering of vertices. In this construction, we show how the problem of assigning vertices and triangles to groups reduces to a well-known NP-hard optimization problem, and present a simple yet effective heuristic solution that performs well in practice. Our array-based Grouper representation also doubles as a triangle mesh data structure that allows direct access to vertices and triangles. Storing only about two integer references per triangle-i.e., less than the three vertex references stored with each triangle in a conventional indexed mesh format-Grouper answers both incidence and adjacency queries in amortized constant time. Our compact representation enables data-parallel processing on multicore computers, instant partitioning and fast transmission for distributed processing, as well as efficient out-of-core access. We demonstrate the versatility and performance benefits of Grouper using a suite of example meshes and processing kernels.

97 MATHEMATICS AND COMPUTING↗

Scalable Contour Tree Computation by Data Parallel Peak Pruning

As data sets grow to exascale, automated data analysis and visualization are increasingly important, to intermediate human understanding and to reduce demands on disk storage via in situ analysis. Trends in architecture of high performance computing systems necessitate analysis algorithms to make effective use of combinations of massively multicore and distributed systems. One of the principal analytic tools is the contour tree, which analyses relationships between contours to identify features of more than local importance. Unfortunately, the predominant algorithms for computing the contour tree are explicitly serial, and founded on serial metaphors, which has limited the scalability of this form of analysis. While there is some work on distributed contour tree computation, and separately on hybrid GPU-CPU computation, there is no efficient algorithm with strong formal guarantees on performance allied with fast practical performance. Furthermore, we report the first shared SMP algorithm for fully parallel contour tree computation, with formal guarantees of O(lg V lg t) parallel steps and O(V lg V) work for data with V samples and t contour tree supernodes, and implementations with more than 30× parallel speed up on both CPU using TBB and GPU using Thrust and up 70× speed up compared to the serial sweep and merge algorithm.

97 MATHEMATICS AND COMPUTING↗

Towards Achieving Performance Portability Using Directives for Accelerators

In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types of node architectures with attached accelerators, both self-hosted multicore and offload multicore/GPU. Our goal is to examine how successful OpenACC and the newer offload features of OpenMP 4.5 are for moving codes between architectures, how much tuning might be required and what lessons we can learn from this experience. To do this, we use examples of algorithms with varying computational intensities for our evaluation, as both compute and data access efficiency are important considerations for overall application performance. We implement these kernels using various methods provided by newer OpenACC and OpenMP implementations, and we evaluate their performance on various platforms including both X86_64 with attached NVIDIA GPUs, self-hosted Intel Xeon Phi KNL, as well as an X86_64 host system with Intel Xeon Phi coprocessors. In this paper, we explain what factors affected the performance portability such as how to pick the right programming model, its programming style, its availability on different platforms, and how well compilers can optimize and target to multiple platforms.

97 MATHEMATICS AND COMPUTING↗

SU-F-BRD-02: Application of ARCHERRT-- A GPU-Based Monte Carlo Dose Engine for Radiation Therapy -- to Tomotherapy and Patient-Independent IMRT

Purpose: As a module of ARCHER -- Accelerated Radiation-transport Computations in Heterogeneous EnviRonments, ARCHER{sub RT} is designed for RadioTherapy (RT) dose calculation. This paper describes the application of ARCHERRT on patient-dependent TomoTherapy and patient-independent IMRT. It also conducts a 'fair' comparison of different GPUs and multicore CPU. Methods: The source input used for patient-dependent TomoTherapy is phase space file (PSF) generated from optimized plan. For patient-independent IMRT, the open filed PSF is used for different cases. The intensity modulation is simulated by fluence map. The GEANT4 code is used as benchmark. DVH and gamma index test are employed to evaluate the accuracy of ARCHER{sub RT} code. Some previous studies reported misleading speedups by comparing GPU code with serial CPU code. To perform a fairer comparison, we write multi-thread code with OpenMP to fully exploit computing potential of CPU. The hardware involved in this study are a 6-core Intel E5-2620 CPU and 6 NVIDIA M2090 GPUs, a K20 GPU and a K40 GPU. Results: Dosimetric results from ARCHER{sub RT} and GEANT4 show good agreement. The 2%/2mm gamma test pass rates for different clinical cases are 97.2% to 99.7%. A single M2090 GPU needs 50~79 seconds for the simulation to achieve a statistical error of 1% in the PTV. The K40 card is about 1.7∼1.8 times faster than M2090 card. Using 6 M2090 card, the simulation can be finished in about 10 seconds. For comparison, Intel E5-2620 needs 507∼879 seconds for the same simulation. Conclusion: We successfully applied ARCHER{sub RT} to Tomotherapy and patient-independent IMRT, and conducted a fair comparison between GPU and CPU performance. The ARCHER{sub RT} code is both accurate and efficient and may be used towards clinical applications.

60 APPLIED LIFE SCIENCES↗

SU-E-T-157: CARMEN: A MatLab-Based Research Platform for Monte Carlo Treatment Planning (MCTP) and Customized System for Planning Evaluation

Purpose: Although there exist several radiotherapy research platforms, such as: CERR, the most widely used and referenced; SlicerRT, which allows treatment plan comparison from various sources; and MMCTP, a full MCTP system; it is still needed a full MCTP toolset that provides users complete control of calculation grids, interpolation methods and filters in order to “fairly” compare results from different TPSs, supporting verification with experimental measurements. Methods: This work presents CARMEN, a MatLab-based platform including multicore and GPGPU accelerated functions for loading RT data; designing treatment plans; and evaluating dose matrices and experimental data.CARMEN supports anatomic and functional imaging in DICOM format, as well as RTSTRUCT, RTPLAN and RTDOSE. Besides, it contains numerous tools to accomplish the MCTP process, managing egs4phant and phase space files.CARMEN planning mode assist in designing IMRT, VMAT and MERT treatments via both inverse and direct optimization. The evaluation mode contains a comprehensive toolset (e.g. 2D/3D gamma evaluation, difference matrices, profiles, DVH, etc.) to compare datasets from commercial TPS, MC simulations (i.e. 3ddose) and radiochromic film in a user-controlled manner. Results: CARMEN has been validated against commercial RTPs and well-established evaluation tools, showing coherent behavior of its multiple algorithms. Furthermore, CARMEN platform has been used to generate competitive complex treatment that has been published in comparative studies. Conclusion: A new research oriented MCTP platform with a customized validation toolset has been presented. Despite of being coded with a high-level programming language, CARMEN is agile due to the use of parallel algorithms. The wide-spread use of MatLab provides straightforward access to CARMEN’s algorithms to most researchers. Similarly, our platform can benefit from the MatLab community scientific developments as filters, registration algorithms etc. Finally, CARMEN arises the importance of grid and filtering control in treatment plan comparison.

60 APPLIED LIFE SCIENCES↗

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↗

Scalable Work Stealing

Irregular and dynamic parallel applications pose significant challenges to achieving scalable performance on large-scale multicore clusters. These applications often require ongoing, dynamic load balancing in order to maintain efficiency. While effective at small scale, centralized load balancing schemes quickly become a bottleneck on large-scale clusters. Work stealing is a popular approach to distributed dynamic load balancing; however its performance on large-scale clusters is not well understood. Prior work on work stealing has largely focused on shared memory machines. In this work we investigate the design and scalability of work stealing on modern distributed memory systems. We demonstrate high efficiency and low overhead when scaling to 8,192 processors for three benchmark codes: a producer-consumer benchmark, the unbalanced tree search benchmark, and a multiresolution analysis kernel.

99 GENERAL AND MISCELLANEOUS↗

Critical Path-Based Thread Placement for NUMA Systems

Multicore multiprocessors use a Non Uniform Memory Architecture (NUMA) to improve their scalability. However, NUMA introduces performance penalties due to remote memory accesses. Without efficiently managing data layout and thread mapping to cores, scientific applications, even if they are optimized for NUMA, may suffer performance loss. In this paper, we present algorithms and a runtime system that optimize the execution of OpenMP applications on NUMA architectures. By collecting information from hardware counters, the runtime system directs thread placement and reduces performance penalties by minimizing the critical path of OpenMP parallel regions. The runtime system uses a scalable algorithm that derives placement decisions with negligible overhead. We evaluate our algorithms and runtime system with four NPB applications implemented in OpenMP. On average the algorithms achieve between 8.13% and 25.68% performance improvement compared to the default Linux thread placement scheme. The algorithms miss the optimal thread placement in only 8.9% of the cases.

97 MATHEMATICS AND COMPUTING↗

A Massively Parallel Adaptive Fast Multipole Method on Heterogeneous Architectures

We describe a parallel fast multipole method (FMM) for highly nonuniform distributions of particles. We employ both distributed memory parallelism (via MPI) and shared memory parallelism (via OpenMP and GPU acceleration) to rapidly evaluate two-body nonoscillatory potentials in three dimensions on heterogeneous high performance computing architectures. We have performed scalability tests with up to 30 billion particles on 196,608 cores on the AMD/CRAY-based Jaguar system at ORNL. On a GPU-enabled system (NSF's Keeneland at Georgia Tech/ORNL), we observed 30x speedup over a single core CPU and 7x speedup over a multicore CPU implementation. By combining GPUs with MPI, we achieve less than 10 ns/particle and six digits of accuracy for a run with 48 million nonuniformly distributed particles on 192 GPUs.

99 GENERAL AND MISCELLANEOUS↗

Critical Path-Based Thread Placement for NUMA Systems

Multicore multiprocessors use Non Uniform Memory Architecture (NUMA) to improve their scalability. However,NUMA introduces performance penalties due to remote memory accesses. Without efficiently managing data layout and thread mapping to cores, scientific applications, even if they are optimized for NUMA, may suffer performance loss. In this paper, we present an algorithm that optimizes the placement of OpenMP threads on NUMA processors. By collecting information from hardware counters and defining new metrics to capture the effects of thread placement, the algorithm reduces NUMA performance penalty by minimizing the critical path of OpenMP parallel regions and by avoiding local memory resource contention. We evaluate our algorithm with NPB benchmarks and achieve performance improvement between 8.13% and 25.68%, compared to the OS default scheduling.

99 GENERAL AND MISCELLANEOUS↗

Investigating applications portability with the Uintah DAG-based runtime system on PetaScale supercomputers

Present trends in high performance computing present formidable challenges for applications code using multicore nodes possibly with accelerators and/or co-processors and reduced memory while still attaining scalability. Software frameworks that execute machine-independent applications code using a runtime system that shields users from architectural complexities offer a possible solution. The Uintah framework for example, solves a broad class of large-scale problems on structured adaptive grids using fluid-flow solvers coupled with particle-based solids methods. Uintah executes directed acyclic graphs of computational tasks with a scalable asynchronous and dynamic runtime system for CPU cores and/or accelerators/co-processors on a node. Uintah's clear separation between application and runtime code has led to scalability increases of 1000x without significant changes to application code. This methodology is tested on three leading Top500 machines; OLCF Titan, TACC Stampede and ALCF Mira using three diverse and challenging applications problems. This investigation of scalability with regard to the different processors and communications performance leads to the overall conclusion that the adaptive DAG-based approach provides a very powerful abstraction for solving challenging multi-scale multi-physics engineering problems on some of the largest and most powerful computers available today.

Meng, Qingyu↗

Towards Batched Linear Solvers on Accelerated Hardware Platforms

As hardware evolves, an increasingly effective approach to develop energy efficient, high-performance solvers, is to design them to work on many small and independent problems. Indeed, many applications already need this functionality, especially for GPUs, which are known to be currently about four to five times more energy efficient than multicore CPUs for every floating-point operation. In this paper, we describe the development of the main one-sided factorizations: LU, QR, and Cholesky; that are needed for a set of small dense matrices to work in parallel. We refer to such algorithms as batched factorizations. Our approach is based on representing the algorithms as a sequence of batched BLAS routines for GPU-contained execution. Note that this is similar in functionality to the LAPACK and the hybrid MAGMA algorithms for large-matrix factorizations. But it is different from a straightforward approach, whereby each of GPU's symmetric multiprocessors factorizes a single problem at a time. We illustrate how our performance analysis together with the profiling and tracing tools guided the development of batched factorizations to achieve up to 2-fold speedup and 3-fold better energy efficiency compared to our highly optimized batched CPU implementations based on the MKL library on a two-sockets, Intel Sandy Bridge server. Compared to a batched LU factorization featured in the NVIDIA's CUBLAS library for GPUs, we achieves up to 2.5-fold speedup on the K40 GPU.

Haidar, Azzam↗

Thread-level parallelization and optimization of NWChem for the Intel MIC architecture

In the multicore era it was possible to exploit the increase in on-chip parallelism by simply running multiple MPI processes per chip. Unfortunately, manycore processors' greatly increased thread- and data-level parallelism coupled with a reduced memory capacity demand an altogether different approach. In this paper we explore augmenting two NWChem modules, triples correction of the CCSD(T) and Fock matrix construction, with OpenMP in order that they might run efficiently on future manycore architectures. As the next NERSC machine will be a self-hosted Intel MIC (Xeon Phi) based supercomputer, we leverage an existing MIC testbed at NERSC to evaluate our experiments. In order to proxy the fact that future MIC machines will not have a host processor, we run all of our experiments in native mode. We found that while straightforward application of OpenMP to the deep loop nests associated with the tensor contractions of CCSD(T) was sufficient in attaining high performance, significant e ort was required to safely and efeciently thread the TEXAS integral package when constructing the Fock matrix. Ultimately, our new MPI+OpenMP hybrid implementations attain up to 65× better performance for the triples part of the CCSD(T) due in large part to the fact that the limited on-card memory limits the existing MPI implementation to a single process per card. Additionally, we obtain up to 1.6× better performance on Fock matrix constructions when compared with the best MPI implementations running multiple processes per card.

CCSD(T)↗

Optimizing Irregular Applications for Energy and Performance on the Tilera Many-core Architecture

Optimizing applications simultaneously for energy and performance is a complex problem. High performance, parallel, irregular applications are notoriously hard to optimize due to their data-dependent memory accesses, lack of structured locality and complex data structures and code patterns. Irregular kernels are growing in importance in applications such as machine learning, graph analytics and combinatorial scientific computing. Performance- and energy-efficient implementation of these kernels on modern, energy efficient, multicore and many-core platforms is therefore an important and challenging problem. We present results from optimizing two irregular applications { the Louvain method for community detection (Grappolo), and high-performance conjugate gradient (HPCCG) { on the Tilera many-core system. We have significantly extended MIT's OpenTuner auto-tuning framework to conduct a detailed study of platform-independent and platform-specific optimizations to improve performance as well as reduce total energy consumption. We explore the optimization design space along three dimensions: memory layout schemes, compiler-based code transformations, and optimization of parallel loop schedules. Using auto-tuning, we demonstrate whole node energy savings of up to 41% relative to a baseline instantiation, and up to 31% relative to manually optimized variants.

irregular applications, energy opttimization, many↗

Damaris: Addressing performance variability in data management for post-petascale simulations

With exascale computing on the horizon, reducing performance variability in data management tasks (storage, visualization, analysis, etc.) is becoming a key challenge in sustaining high performance. Here, this variability significantly impacts the overall application performance at scale and its predictability over time. In this article, we present Damaris, a system that leverages dedicated cores in multicore nodes to offload data management tasks, including I/O, data compression, scheduling of data movements, in situ analysis, and visualization. We evaluate Damaris with the CM1 atmospheric simulation and the Nek5000 computational fluid dynamic simulation on four platforms, including NICS’s Kraken and NCSA’s Blue Waters. Our results show that (1) Damaris fully hides the I/O variability as well as all I/O-related costs, thus making simulation performance predictable; (2) it increases the sustained write throughput by a factor of up to 15 compared with standard I/O approaches; (3) it allows almost perfect scalability of the simulation up to over 9,000 cores, as opposed to state-of-the-art approaches that fail to scale; and (4) it enables a seamless connection to the VisIt visualization software to perform in situ analysis and visualization in a way that impacts neither the performance of the simulation nor its variability. In addition, we extended our implementation of Damaris to also support the use of dedicated nodes and conducted a thorough comparison of the two approaches—dedicated cores and dedicated nodes—for I/O tasks with the aforementioned applications.

97 MATHEMATICS AND COMPUTING↗

Implementing Arbitrary/Common Concurrent Writes of CRCW PRAM

The Parallel Random Access Machines (PRAM) abstraction is the simplest and most elegant algorithmic model for the design and analysis of parallel algorithms. It consists of different models categorized based on the underlying memory access mode used, the most powerful of which is the Concurrent Read Concurrent Write (CRCW) model. A PRAM algorithm describes a series of rounds, each of which consists of a collection of operations that can be executed concurrently within the same time step. However, the lack of support for concurrent memory accesses and the prevalence of asynchronous programming models led to the belief that implementing CRCW PRAM algorithms is unattainable and prompted many to avoid this model except for theoretical studies of optimal performance.In this work, we study the arbitrary and common concurrent writes in the CRCW PRAM model and explore implementation challenges on general-purpose systems. Moreover, we examine current practices for implementing common/arbitrary concurrent writes and propose a new efficient lightweight and thread-safe method to implement concurrent writes through leveraging atomic instructions. To demonstrate the efficacy of our method, we developed OpenMP kernels for classical CRCW PRAM algorithms and provide experimental results and comparisons based on run time performance measured over the x86 multicore architecture. Our results show a performance speedup compared to current practices up to 4.5x across all our benchmarks.

Ghanim, Fady↗

Complexity in scalable computing.

The rich history of scalable computing research owes much to a rapid rise in computing platform scale in terms of size and speed. As platforms evolve, so must algorithms and the software expressions of those algorithms. Unbridled growth in scale inevitably leads to complexity. This special issue grapples with two facets of this complexity: scalable execution and scalable development. The former results from efficient programming of novel hardware with increasing numbers of processing units (e.g., cores, processors, threads or processes). The latter results from efficient development of robust, flexible software with increasing numbers of programming units (e.g., procedures, classes, components or developers). The progression in the above two parenthetical lists goes from the lowest levels of abstraction (hardware) to the highest (people). This issue's theme encompasses this entire spectrum. The lead author of each article resides in the Scalable Computing Research and Development Department at Sandia National Laboratories in Livermore, CA. Their co-authors hail from other parts of Sandia, other national laboratories and academia. Their research sponsors include several programs within the Department of Energy's Office of Advanced Scientific Computing Research and its National Nuclear Security Administration, along with Sandia's Laboratory Directed Research and Development program and the Office of Naval Research. The breadth of interests of these authors and their customers reflects in the breadth of applications this issue covers. This article demonstrates how to obtain scalable execution on the increasingly dominant high-performance computing platform: a Linux cluster with multicore chips. The authors describe how deep memory hierarchies necessitate reducing communication overhead by using threads to exploit shared register and cache memory. On a matrix-matrix multiplication problem, they achieve up to 96% parallel efficiency with a three-part strategy: intra-node multithreading, non-blocking inter-node message passing, and a dedicated communications thread to facilitate concurrent communications and computations. On a quantum chemistry problem, they spawn multiple computation threads and communication threads on each node and use one-sided communications between nodes to minimize wait times. They reduce software complexity by evolving a multi-threaded factory pattern in C++ from a working, message-passing program in C.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗