Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “OpenCL”

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 73 records · Page 4

Compiler-based code generation and autotuning for geometric multigrid on GPU-accelerated supercomputers

GPUs, with their high bandwidths and computational capabilities are an increasingly popular target for scientific computing. Unfortunately, to date, harnessing the power of the GPU has required use of a GPU-specific programming model like CUDA, OpenCL, or OpenACC. Thus, in order to deliver portability across CPU-based and GPU-accelerated supercomputers, programmers are forced to write and maintain two versions of their applications or frameworks. In this paper, we explore the use of a compiler-based autotuning framework based on CUDA-CHiLL to deliver not only portability, but also performance portability across CPU- and GPU-accelerated platforms for the geometric multigrid linear solvers found in many scientific applications. We also show that with autotuning we can attain near Roofline (a performance bound for a computation and target architecture) performance across the key operations in the miniGMG benchmark for both CPU- and GPU-based architectures as well as for a multiple stencil discretizations and smoothers. We show that our technology is readily interoperable with MPI resulting in performance at scale equal to that obtained via hand-optimized MPI+CUDA implementation.

97 MATHEMATICS AND COMPUTING↗

Toward performance-portable PETSc for GPU-based exascale systems

The Portable Extensible Toolkit for Scientific computation (PETSc) library delivers scalable solvers for nonlinear time-dependent differential and algebraic equations and for numerical optimization. The PETSc design for performance portability addresses fundamental GPU accelerator challenges and stresses flexibility and extensibility by separating the programming model used by the application from that used by the library, and it enables application developers to use their preferred programming model, such as Kokkos, RAJA, SYCL, HIP, CUDA, or OpenCL, on upcoming exascale systems. Furthermore, a blueprint for using GPUs from PETSc-based codes is provided, and case studies emphasize the flexibility and high performance achieved on current GPU-based systems.

97 MATHEMATICS AND COMPUTING↗

High Energy Electromagnetic Particle Transportation on the GPU

We present massively parallel high energy electromagnetic particle transportation through a finely segmented detector on a Graphics Processing Unit (GPU). Simulating events of energetic particle decay in a general-purpose high energy physics (HEP) detector requires intensive computing resources, due to the complexity of the geometry as well as physics processes applied to particles copiously produced by primary collisions and secondary interactions. The recent advent of hardware architectures of many-core or accelerated processors provides the variety of concurrent programming models applicable not only for the high performance parallel computing, but also for the conventional computing intensive application such as the HEP detector simulation. The components of our prototype are a transportation process under a non-uniform magnetic field, geometry navigation with a set of solid shapes and materials, electromagnetic physics processes for electrons and photons, and an interface to a framework that dispatches bundles of tracks in a highly vectorized manner optimizing for spatial locality and throughput. Core algorithms and methods are excerpted from the Geant4 toolkit, and are modified and optimized for the GPU application. Program kernels written in C/C++ are designed to be compatible with CUDA and OpenCL and with the aim to be generic enough for easy porting to future programming models and hardware architectures. To improve throughput by overlapping data transfers with kernel execution, multiple CUDA streams are used. Issues with floating point accuracy, random numbers generation, data structure, kernel divergences and register spills are also considered. Performance evaluation for the relative speedup compared to the corresponding sequential execution on CPU is presented as well.

Canal, P.↗

IRIS: A Portable Runtime System Exploiting Multiple Heterogeneous Programming Systems

Across embedded, mobile, enterprise, and HPC systems, computer architectures are becoming more heterogeneous and complex. 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 specialize for each architecture. As we show, all of these approaches critically depend on their runtime system for discovery, execution, scheduling, and data orchestration. To address this challenge, we believe that a more agile and proactive runtime system is essential to increase performance portability and improve user productivity. In this regard, we have designed and implemented IRIS: a portable runtime system exploiting multiple heterogeneous programming systems. IRIS can discover available resources, manage multiple diverse programming systems (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. 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.

Kim, Jungwon↗

IRIS-DMEM: Efficient Memory Management for Heterogeneous Computing

This paper proposes an efficient data memory management approach for the Intelligent RuntIme System (IRIS) heterogeneous computing framework along with new data transfer policies. IRIS provides a task-based programming model for extreme heterogeneous computing (e.g., CPU, GPU, DSP, FPGA) with support for today's most important programming languages (e.g., OpenMP, OpenCL, CUDA, HIP, OpenACC). However, the IRIS framework either forces the programmer to introduce data transfer commands for each task or relies on suboptimal memory management for automatic and transparent data transfers. The work described here extends IRIS with novel heterogeneous memory handling and introduces novel data transfer policies by employing the Distributed data MEMory handler (DMEM) for efficient and optimal movement of data among the various computing resources. The proposed approach achieves performance gains of up to 7× for tiled LU factorization and tiled DGEMM (i.e., matrix multiplication) benchmarks. Moreover, this approach also reduces data transfers by up to 71% when compared to previous IRIS heterogeneous memory management handlers. This work compares the performance results of the IRIS framework's novel DMEM with the StarPU runtime and MAGMA math library for GPUs. Experiments show a performance gain of up to 1.95× over StarPU and 2.1× over MAGMA.

Miniskar, Narasinga Rao↗

Implementing the OpenACC Data Model

Programming accelerators today usually requires managing separate virtual and physical memories, such as allocating space in and copying data between host and device memories. The OpenACC API provides data directives and clauses to control this behavior where it is required. This paper describes how the data model is supported in current OpenACC implementations, ranging from research compilers (OpenUH and OpenARC) to a commercial compiler (the PGI OpenACC compiler). This includes implementation of the data directives and clauses, testing whether the data is already present on the device, OpenCL support, managing asynchronous data transfers and memory allocations, handling aliased data, reusing device memory, managing partially present data, and support for shared memory between host and device. Lastly, it also discusses on-going work to manage large, complex dynamic data structures.

Wolfe, Michael↗

Snowflake: A Lightweight Portable Stencil DSL

Stencil computations are not well optimized by general-purpose production compilers and the increased use of multicore, manycore, and accelerator-based systems makes the optimization problem even more challenging. In this paper we present Snowflake, a Domain Specific Language (DSL) for stencils that uses a 'micro-compiler' approach, i.e., small, focused, domain-specific code generators. The approach is similar to that used in image processing stencils, but Snowflake handles the much more complex stencils that arise in scientific computing, including complex boundary conditions, higher-order operators (larger stencils), higher dimensions, variable coefficients, non-unit-stride iteration spaces, and multiple input or output meshes. Snowflake is embedded in the Python language, allowing it to interoperate with popular scientific tools like SciPy and iPython; it also takes advantage of built-in Python libraries for powerful dependence analysis as part of a just-in-time compiler. We demonstrate the power of the Snowflake language and the micro-compiler approach with a complex scientific benchmark, HPGMG, that exercises the generality of stencil support in Snowflake. By generating OpenMP comparable to, and OpenCL within a factor of 2x of hand-optimized HPGMG, Snowflake demonstrates that a micro-compiler can support diverse processor architectures and is performance-competitive whilst preserving a high-level Python implementation.

97 MATHEMATICS AND COMPUTING↗

Evaluating CUDA Portability with HIPCL and DPCT

HIPCL is expanding the scope of the CUDA portability route from an AMD platform to an OpenCL platform. In the meantime, the Intel DPC++ Compatibility Tool (DPCT) is migrating a CUDA program to a data parallel C++ (DPC++) program. Towards the goal of portability enhancement, we evaluate the performance of the CUDA applications from Rodinia, SHOC, and proxy applications ported using HIPCL and DPCT on Intel GPUs. After profiling the ported programs, we aim to understand their performance gaps, and optimize codes converted by DPCT to improve their performance. The open-source repository for the CUDA, HIP, and DPCT programs will be useful for the development of a translator.

Jin, Zheming↗

A Task Based Approach for Co-Scheduling Ensemble Workloads on Heterogeneous Nodes

Scientific workflows consist of multiple, connected applications, with data and results flowing from one to another in a pipeline. Traditionally, such workflows are executed in sequential order, storing intermediate data in storage disks. Co-scheduling application workflows concurrently on the same compute nodes would greatly reduce the cost of moving data to/from storage and allow real-time analysis of intermediate results. Nevertheless, most parallel programming runtimes do not allow seamless integration of various applications in a scientific workflow, in part due to the complexity of managing data and resources. The situation is even more complicated for heterogeneous systems. In this work we extend the Minos Computing Library (MCL) runtime to accelerate pipe-lined and parallel workloads where multiple applications are running in the same system. MCL’s asynchronous task library and runtime dynamically manages resources to allow co-scheduling of multiple processes sharing heterogeneous resources. In addition, we design a custom ex- tension of the Open Compute Language (OpenCL) to enable multiple processes to share device memory. We enable MCL to coordinate these shared buffers to allow for easy, fast data sharing between applications. Using malleable micro-benchmarks and two application workflows that combine scientific simulation and AI-based analysis, we show that our method outperforms traditional approaches.

Index Terms—Parallel systems, Scheduling and Task ↗

IRIS: Exploring Performance Scaling of the Intelligent Runtime System and its Dynamic Scheduling Policies

High-Performance Computing is becoming increasingly heterogeneous, relying on a diverse mix of hardware to achieve good performance. Paradoxically, current drivers and frameworks for these devices typically require separate languages and implementations for each vendor. Furthermore, there are few tools and little support to schedule codes between these devices in a truly heterogeneous manner-partly because of this fragmentation between vendors and the languages each supports. To overcome both limitations, the Intelligent Runtime System (IRIS) was developed. It allows a common task abstraction to automatically be shared among contemporary vendors and is run from a single host-side API. At runtime, IRIS queries the host system and registers which frameworks and drivers are available, these determine which kernels can be used by the scheduler-CPUs via OpenMP, Nvidia GPUs (CUDA), AMD GPUs (HIP), and Intel and Xilinx FPGAs with OpenCL. IRIS enables tasks to be scheduled to any heterogeneous device and resolves to the appropriate kernel binary at runtimeit only uses the devices supported by the system on which it is run. IRIS supports single-task and graph-based expressions of dependencies of tasks. Additionally, IRIS features a range of dynamic scheduling policies, allowing complex chains of tasks and interactions to be executed, relieving the programmer/user from considering the system to assign tasks to devices optimally. This paper presents the peak performance attainable by IRIS over a range of systems-each with different numbers and types of accelerator devices, it highlights the flexibility of IRIS since these devices are truly heterogeneous, relying on different backends (drivers, frameworks, and languages) which historically required unique implementations to utilize them. We then use this peak performance as a baseline to compare increasingly complex chains of tasks (with increasingly complex task dependencies) and evaluate how IRIS copes. Finally, we consider the performance of different IRIS scheduling policies on this range of task graphs.

Johnston, Beau↗

Case Study of Using Kokkos and SYCLs Performance-Portable Frameworks for Milc-Dslash Benchmark on NVIDIA, AMD and Intel GPUs

Six of the top ten supercomputers in the TOP500 list from June 2021 rely on NVIDIA GPUs to achieve their peak compute bandwidth. With the announcement of Aurora, Frontier, and El Capitan, Intel and AMD have also entered the domain of providing GPUs for scientific computing. A consequence of the increased diversity in the GPU landscape is the emergence of portable programming models such as Kokkos, SYCL, OpenCL, and OpenMP, which allow application developers to maintain a single-source code across a diverse range of hardware architectures. While the portable frameworks try to optimize the compute resource usage on a given architecture, it is the programmers responsibility to expose parallelism in an application that can take advantage of thousands of processing elements available on GPUs. In this paper, we introduce a GPU-friendly parallel implementation of Milc-Dslash that exposes multiple hierarchies of parallelism in the algorithm. Milc-Dslash was designed to serve as a benchmark with highly optimized matrix-vector multiplications to measure the resource utilization on the GPU systems. The parallel hierarchies in the Milc-Dslash algorithm are mapped onto a target hardware using Kokkos and SYCL programming models. We present the performance achieved by Kokkos and SYCL implementations of Milc-Dslash on NVIDIA A100 GPU, AMD MI100 GPU, and Intel Gen9 GPU. Additionally, we compare the Kokkos and SYCL performances with those obtained from the versions written in CUDA and HIP programming models on NVIDIA A100 GPU and AMD MI100 GPU, respectively.

Dufek, Amanda S↗

Leveraging Compiler-Based Translation to Evaluate a Diversity of Exascale Platforms

Accelerator-based heterogeneous computing is the de facto standard in current and upcoming exascale machines. These heterogeneous resources empower computational scientists to select a machine or platform well-suited to their domain or applications. However, this diversity of machines also poses challenges related to programming model selection: inconsistent availability of programming models across different exascale systems, lack of performance portability for those programming models that do span several systems, and inconsistent performance between different models on a single platform. We explore these challenges on exascale-similar hardware, including AMD MI100 and NVIDIA A100 GPUs. By extending the sourceto-source compiler OpenARC, we demonstrate the power of automated translation of applications written in a single frontend programming model (OpenACC) into a variety of backend models (OpenMP, OpenCL, CUDA, HIP) that span the upcoming exascale environments. This translation enables us to compare performance within and across devices and to analyze programming model behavior with profiling tools.

Lambert, Jacob↗

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↗

CONRAD—A software framework for cone-beam imaging in radiology

Purpose: In the community of x-ray imaging, there is a multitude of tools and applications that are used in scientific practice. Many of these tools are proprietary and can only be used within a certain lab. Often the same algorithm is implemented multiple times by different groups in order to enable comparison. In an effort to tackle this problem, the authors created CONRAD, a software framework that provides many of the tools that are required to simulate basic processes in x-ray imaging and perform image reconstruction with consideration of nonlinear physical effects.Methods: CONRAD is a Java-based state-of-the-art software platform with extensive documentation. It is based on platform-independent technologies. Special libraries offer access to hardware acceleration such as OpenCL. There is an easy-to-use interface for parallel processing. The software package includes different simulation tools that are able to generate up to 4D projection and volume data and respective vector motion fields. Well known reconstruction algorithms such as FBP, DBP, and ART are included. All algorithms in the package are referenced to a scientific source.Results: A total of 13 different phantoms and 30 processing steps have already been integrated into the platform at the time of writing. The platform comprises 74.000 nonblank lines of code out of which 19% are used for documentation. The software package is available for download at http://conrad.stanford.edu. To demonstrate the use of the package, the authors reconstructed images from two different scanners, a table top system and a clinical C-arm system. Runtimes were evaluated using the RabbitCT platform and demonstrate state-of-the-art runtimes with 2.5 s for the 256 problem size and 12.4 s for the 512 problem size.Conclusions: As a common software framework, CONRAD enables the medical physics community to share algorithms and develop new ideas. In particular this offers new opportunities for scientific collaboration and quantitative performance comparison between the methods of different groups.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

SU-E-T-423: Fast Photon Convolution Calculation with a 3D-Ideal Kernel On the GPU

Purpose: The calculation time is a trade-off for improving the accuracy of convolution dose calculation with fine calculation spacing of the KERMA kernel. We investigated to accelerate the convolution calculation using an ideal kernel on the Graphic Processing Units (GPU). Methods: The calculation was performed on the AMD graphics hardware of Dual FirePro D700 and our algorithm was implemented using the Aparapi that convert Java bytecode to OpenCL. The process of dose calculation was separated with the TERMA and KERMA steps. The dose deposited at the coordinate (x, y, z) was determined in the process. In the dose calculation running on the central processing unit (CPU) of Intel Xeon E5, the calculation loops were performed for all calculation points. On the GPU computation, all of the calculation processes for the points were sent to the GPU and the multi-thread computation was done. In this study, the dose calculation was performed in a water equivalent homogeneous phantom with 150{sup 3} voxels (2 mm calculation grid) and the calculation speed on the GPU to that on the CPU and the accuracy of PDD were compared. Results: The calculation time for the GPU and the CPU were 3.3 sec and 4.4 hour, respectively. The calculation speed for the GPU was 4800 times faster than that for the CPU. The PDD curve for the GPU was perfectly matched to that for the CPU. Conclusion: The convolution calculation with the ideal kernel on the GPU was clinically acceptable for time and may be more accurate in an inhomogeneous region. Intensity modulated arc therapy needs dose calculations for different gantry angles at many control points. Thus, it would be more practical that the kernel uses a coarse spacing technique if the calculation is faster while keeping the similar accuracy to a current treatment planning system.

60 APPLIED LIFE SCIENCES↗

SU-D-206-01: Employing a Novel Consensus Optimization Strategy to Achieve Iterative Cone Beam CT Reconstruction On a Multi-GPU Platform

Purpose: While compressed sensing-based cone-beam CT (CBCT) iterative reconstruction techniques have demonstrated tremendous capability of reconstructing high-quality images from undersampled noisy data, its long computation time still hinders wide application in routine clinic. The purpose of this study is to develop a reconstruction framework that employs modern consensus optimization techniques to achieve CBCT reconstruction on a multi-GPU platform for improved computational efficiency. Methods: Total projection data were evenly distributed to multiple GPUs. Each GPU performed reconstruction using its own projection data with a conventional total variation regularization approach to ensure image quality. In addition, the solutions from GPUs were subject to a consistency constraint that they should be identical. We solved the optimization problem with all the constraints considered rigorously using an alternating direction method of multipliers (ADMM) algorithm. The reconstruction framework was implemented using OpenCL on a platform with two Nvidia GTX590 GPU cards, each with two GPUs. We studied the performance of our method and demonstrated its advantages through a simulation case with a NCAT phantom and an experimental case with a Catphan phantom. Result: Compared with the CBCT images reconstructed using conventional FDK method with full projection datasets, our proposed method achieved comparable image quality with about one third projection numbers. The computation time on the multi-GPU platform was ∼55 s and ∼ 35 s in the two cases respectively, achieving a speedup factor of ∼ 3.0 compared with single GPU reconstruction. Conclusion: We have developed a consensus ADMM-based CBCT reconstruction method which enabled performing reconstruction on a multi-GPU platform. The achieved efficiency made this method clinically attractive.

60 APPLIED LIFE SCIENCES↗

Fast Mesh-to-Mesh Remaps Using Hash Algorithms

Here, we explore the potential uses of hash-based algorithms in the remap operation, mapping one computational mesh onto another. We implement and test optimizations designed to reduce memory operations and compare these optimizations in order to evaluate the performance impact across different meshes. In particular, we created a new remap method that creates a structured hierarchy of hashes. Tests were run on 2D meshes on up to 14 levels of mesh refinement and hundreds of millions of cells to measure the algorithms' relative speeds under differing circumstances. We show that our methods perform two orders of magnitude better than the standard kD-tree algorithm and that on different meshes different hash methods can be optimal. We explore these algorithms on serial CPUs, GPUs using OpenCL, and multi-core CPUs using OpenMP.

97 MATHEMATICS AND COMPUTING↗