Meeting your Performance Portability goals with RAJA : A case study using CoMD – a proxy application for molecular dynamics. C/C++ with RAJA
Explore the source record for details and available documents.
SEARCH · Search NASA
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.
Explore the source record for details and available documents.
The RAJA Performance Suite is designed to evaluate performance of the RAJA performance portability library on a wide variety of important high performance computing (HPC) algorithmic lulmels. These kernels assess compiler optimizations and various parallel programming model backends accessible through RAJA, such as OpenMP, CUDA, etc. The Initial version of the suite contains 25 computational kernels, each of which appears in 6 variants: Baseline SequcntiaJ, RAJA SequentiaJ, Baseline OpenMP, RAJA OpenMP, Baseline CUDA, RAJA CUDA. All variants of each kernel perform essentially the same mathematical operations and the loop body code for each kernel is identical across all variants. There are a few kernels, such as those that contain reduction operations, that require CUDA-specific coding for their CUDA variants. ActuaJ computer instructions executed and how they run in parallel differs depending on the parallel programming model backend used and which optimizations are perfonned by the compiler used to build the Perfonnance Suite executable. The Suite will be used primarily by RAJA developers to perform regular assessments of RAJA performance across a range of hardware platforms and compilers as RAJA features are being developed. It will also be used by LLNL hardware and software vendor panners for new defining requirements for future computing platform procurements and acceptance testing. In particular, the RAJA Performance Suite will be used for compiler acceptance testing of the upcoming CORAUSierra machine {initial LLNL delivery expected in late-2017/early 2018) and the CORAL-2 procurement. The Suite will aJso be used to generate concise source code reproducers of compiler and runtime issues we uncover so that we may provide them to relevant vendors to be fixed.
As computer architectures become increasingly complex and diverse, application developers face difficult challenges to achieve high performance while maintaining code portability. The problem is especially acute for large ASC multiphysics codes. Efficient parallel execution often requires tuning algorithms and data access to match processor and memory system constraints. Changing compiler directives and parallel programming model constructs on thousands of individual loops in a large code is disruptive and unwieldy. RAJA is a programming approach that we have been developing at Lawrence Livermore National Laboratory to encapsulate platform-specific concerns, related to both hardware and parallel programming models. The RAJA abstraction layer simplifies porting C/C++ codes to various programming models and architectures by reducing effort and developer disruption. In this report, we motivate and describe key aspects of RAJA. We also present a preliminary assessment of RAJA based on exploration in three ASC hydrodynamics codes at LLNL, which was one part of a three-part ASC Level 2 milestone, completed in September 2014.
RAJA is a collection of C++ software abstractions designed to make HPC applications portable across a range of hardware architectures and programming models. The main conceptual abstraction in FAJA is a loop. RAJA provides abstractions for types, loop execution and scheduling, and loop iteration space portioning and ordering cooperate to insulate algorithms in application code that domain experts write from implementations details of those algorithms that may need to depend on the underlying computer hardware for high performance.
Modern high-performance computing architectures are diverse, with hardware designs that feature multicore CPUs and CPUs plus accelerators, such as GPUs or FPGAs. Achieving good application performance often requires using a programming model and managing application data in ways that are best suited to a particular platform. For large codes that are used daily in production, and are under continual development, architecture-specific ports are untenable. Maintainability requires single-source code that is performance portable across a range of architectures and programming model back-ends. This report describes RAJA, Umpire, and CHAI, three software libraries that provide complementary capabilities to enable application portability for high performance computing (HPC) platforms. In particular, these libraries provide software abstractions that insulate application source code from hardware architecture and programming model-specific implementation details. These development of these libraries was motivated by the needs of multiphysics applications in the Weapons Simulation and Computing Program (WSC) at Lawrence Livermore National Laboratory (LLNL). Funding though the Exascale Computing Project (ECP) Software Technology (ST) project 2.3.1.06 RAJA has fostered outreach and collaboration with the ECP application and software technology development communities. In this report, we describe ECP adoption and integration of these tools to this point. The larger goal of the report is to describe to the ECP community what these projects offer, how they are used, and how to approach adoption and integration of them.
RAJA is a collection of C++ software abstractions designed to make HPC applications portable across a range of hardware architectures and programming models. The main conceptual abstraction in FAJA is a loop. RAJA provides abstractions for types, loop execution and scheduling, and loop iteration space portioning and ordering cooperate to insulate algorithms in application code that domain experts write from implementations details of those algorithms that may need to depend on the underlying computer hardware for high performance.
CHAI and RAJA provide an excellent base on which to build portable codes. CARE expands that functionality, adding new features such as loop fusion capability and a portable interface for many numerical algorithms. It provides all the basics for anyone wanting to write portable code.
Nearly 120 years ago, the great “San Francisco” earthquake of 1906 provided a stark and sobering view of the havoc that can be caused by the sudden and violent movement of Earth’s tectonic plates. According to USGS, the rupture along the San Andreas fault extended 296 miles (447 kilometers) and shook so violently that the motion could be felt as far north as Oregon and east into Nevada. The estimated 7.9-magnitude quake and subsequent fires decimated the major metropolis and surrounding areas: buildings turned to ruins, hundreds of thousands of people left homeless, and a death toll exceeding 3,000. Today, as evidenced by the catastrophic 7.8-magnitude earthquake that struck Turkey in February of 2023, large earthquakes still present a significant potential danger to life and economic security as researchers work to develop ways to better understand earthquake phenomena and quantify associated risks.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
In an effort to understand the opaque nature of high-voltage nanosecond microdischarges and their ability to block 532 nm nanosecond laser pulses, as demonstrated by Bataller et al. [Appl. Phys. Lett. 105, 223501 (2014)], Levko and Raja have published a paper [Phys. Plasmas 22, 123518 (2015)] which simulates spark discharges in 1 bar of xenon gas. At this ambient pressure, Levko and Raja simulate final electron densities much too low to explain the observed opacity and conclude the probing laser generates the requisite ionization for self-blocking. However, the experimental findings of Bataller et al. showed opacity being reached at 10 bar, an order of magnitude larger than the pressure simulated by Levko and Raja. Furthermore, Bataller et al. showed the spark remained opaque for very low laser intensities, insufficient to cause any significant ionization. Although the simulation performed by Levko and Raja could have merit at 1 bar, it is not an appropriate comparison to the regime studied by Bataller et al.
Purpose of Review: Acute pulmonary embolism (PE) is the third most-common acute cardiovascular disease in the United States. Nearly 900,000 new diagnoses are made each year, with 250,000 associated hospitalizations and up to 100,000 deaths (Palacio et al. in Semin Roentgenol 50(3):217–225, 2015; Metter et al. in Am J Roentgenol 208(3):489–494, 2017; Jaff et al. in Circulation 123(16):1788–830, 2011). Timely and accurate diagnosis of acute PE is crucial for the prevention of most deaths associated with PE, even more so than optimal medical therapy (Palacio et al. in Semin Roentgenol 50(3):217–225, 2015; Jaff et al. in Circulation 123(16):1788–830, 2011; Raja et al. in Ann Intern Med 163(9):701, 2015). Recent Findings: Computed tomography (CT) has emerged as the primary imaging modality used in the diagnosis of acute PE. Thus, radiologists play a key role in the correct diagnosis of acute PE, and must differentiate this diagnosis from other conditions that mimic the imaging findings of acute PE, as well as from artifacts associated with imaging techniques (Palacio et al. in Semin Roentgenol 50(3):217–225, 2015; Jaff et al. in Circulation 123(16):1788–830, 2011; Raja et al. in Ann Intern Med 163(9):701, 2015). Understanding how these diagnoses present on CT is necessary for improved patient care. Summary: This article will discuss the diagnosis of acute PE with CT, the diagnosis of diseases that may be mistaken for acute PE, and how to identify imaging artifacts that may be mistaken for an acute PE.
Seismic waves fourth order (SW4) solves the seismic wave equations on Cartesian and curvilinear grids using large compute clusters with O (100,000) cores. In this article we discuss the porting of SW4 to run on the CORAL architecture using the RAJA performance portability abstraction layer. The performances of key kernels using RAJA and CUDA are compared to estimate the performance penalty of using the portability abstraction layer. Code changes required for efficiency on GPUs and minimizing time spent in Message Passing Interface (MPI) are discussed. This article describes a path for efficiently porting large code bases to GPU-based machines while avoiding the pitfalls of a new architecture in the early stages of its deployment. Current bottlenecks in the code are discussed along with possible architectural or software mitigations. SW4 runs 28× faster on one 4-GPU CORAL node than on a CTS-1 node (Dual Intel Xeon E5-2695 v4). SW4 is now in routine use on problems of unprecedented resolution (203 billion grid points) and scale on 1,200 nodes of Summit.
CAMP provides a portable optimized implementation of template meta-programming primitives. All capabilities found within are available in other open source projects, including metal and mpl among others, but no other library we know of supports all compilers required by LLNL codes. CAMP serves as an efficient facility for writing portable compile-time operations in C++ codes. It was developed as part of the RAJA project to facilitate more complex compile-time introspection and better error message delivery to users, but has since proven useful in Umpire, CHAI and other pre-release LLNL software projects and is thus being re-factored to stand alone. The CAMP library uses portable C++ template mechanisms almost exclusively, with the exception of intrinsics provided for performance by some compilers such as clang and gee. There are no private or protected interfaces and no proprietary information included in the source. Any and all future versions will limit themselves to implementing commonly available and well known primitives for compile-time manipulation of types. A new public github repository will be created to host the code independent of its current location in the raja repository, likely at github.com/llnl/camp.
A 10 MeV, 5 kW S-band travelling wave electron linear accelerator has been developed at Raja Ramanna Centre for Advanced Technology for industrial applications. The electron accelerating structure is made of oxygen-free electronic (OFE) copper and comprises of fifty-six brazed joints and is fabricated in multiple stages, as per the process and testing requirements. The accelerating structure is often required to be repaired through an additional brazing cycle with the same filler. The present study evaluates possible microstructural and mechanical property degradation of OFE copper brazed joints, as a result of its exposure to multiple brazing cycles. The study was performed on an OFE copper cavity stack brazed assembly, made through a four-stage brazing using BVAg-8 as braze filler. The resultant brazed assembly did not exhibit any leak at a sensitivity of 1 × 10{sup −10} mbar.L/s. The results of the study demonstrate that OFE copper brazements can be safely subjected to an additional brazing cycle without causing significant degradation in its microstructure and strength. However, brazed joint’s exposure to any additional brazing cycles results in the formation of voids and cracks due to intergranular penetration of silver, with associated drop in tensile strength from about 210 MPa to about 150-180 MPa. Based on the results of study, it is recommended to qualify the accelerating structure, fabricated through multiple brazing cycles, by a suitable non-destructive technique. The results of the study are important for design, fabrication and maintenance of electron linear accelerator components.