Proteus: Portable Runtime Optimization of GPU Kernel Execution with Just-In-Time Compilation
Explore the source record for details and available documents.
Georgakoudis, G.↗
SEARCH · Search NASA
Results for “Compiler Optimizations”
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.
OpenARC
OpenARC is an open-sourced, very High-Level Intermediate Representation (HLIR)-based, extensible compiler framework, where various performance optimizations, traceability mechanisms, fault tolerance techniques, etc., can be built for better debuggability/performance/resilience on the complex accelerator computing. OpenARC is the first OpenACC compiler supporting Altera FPGAs, in addition to NVIDIA GPUs, AMD GPUs, and Intel Xeon Phis.
Lee, Seyong [Oak Ridge National Laboratory (ORNL),↗
Quantum Fast Circuit Optimizer (qFactor) v1
Quantum programs are represented as circuits, or sequences of gates. A common problem during quantum compilation to find a circuit that implements a target program. This is an optimizer, it optimizes a circuit's distance to the target by solving for the best fit gates. It is not a standard optimizer, it only optimizes circuits. It is not a compiler or synthesizer, it cannot build circuits, but given a circuit structure, it will solve for the gates. Quantum compilers (qsearch, qfast) use standard numerical optimizers. This has advantages over those general-purpose optimizers since it is domain-specific.
Younis, Ed↗
Multiprocessors and runtime compilation
Runtime preprocessing plays a major role in many efficient algorithms in computer science, as well as playing an important role in exploiting multiprocessor architectures. Examples are given that elucidate the importance of runtime preprocessing and show how these optimizations can be integrated into compilers. To support the arguments, transformations implemented in prototype multiprocessor compilers are described and benchmarks from the iPSC2/860, the CM-2, and the Encore Multimax/320 are presented.
Saltz, Joel↗
CCAMP: OpenMP and OpenACC Interoperable Framework
Heterogeneous systems have become a staple of the HPC environment. Several directive-based solutions, such as OpenMP and OpenACC, have been developed to alleviate the challenges of programming heterogeneous systems, and these standards strive to provide a single portable programming solution across heterogeneous environments. However, in many ways this goal has yet to be realized due to device-specific implementations and different levels of language support across compilers. In this framework we aim to analyze and address the different levels of optimization and compatibility between OpenACC and OpenMP programs and device compilers. We introduce the CCAMP framework, built on the OpenARC compiler, which implements language translation between OpenACC and OpenMP, with the goal of exploiting the maturity of different device-specific compilers to maximize performance for a given architecture. We show that CCAMP allows us to generate code for a specific device-compiler combination given a device-agnostic OpenMP or OpenACC program, allowing compilation and execution of programs with specific directives on otherwise incompatible devices. CCAMP also provides a starting point for a more advanced interoperable framework that can effectively provide directive translation and device, compiler, and application specific optimizations.
Lambert, Jacob↗
Preparation and Analysis of Platinum Thin Films for High Temperature Sensor Applications
A study has been made of platinum thin films for application as high temperature resistive sensors. To support NASA Glenn Research Center s high temperature thin film sensor effort, a magnetron sputtering system was installed recently in the GRC Microsystems Fabrication Clean Room Facility. Several samples of platinum films were prepared using various system parameters to establish run conditions. These films were characterized with the intended application of being used as resistive sensing elements, either for temperature or strain measurement. The resistances of several patterned sensors were monitored to document the effect of changes in parameters of deposition and annealing. The parameters were optimized for uniformity and intrinsic strain. The evaporation of platinum via oxidation during annealing over 900 C was documented, and a model for the process developed. The film adhesion was explored on films annealed to 1000 C with various bondcoats on fused quartz and alumina. From this compiled data, a list of optimal parameters and characteristics determined for patterned platinum thin films is given.
Wrbanek, John D.↗
Computer-Aided Parallelizer and Optimizer
The Computer-Aided Parallelizer and Optimizer (CAPO) automates the insertion of compiler directives (see figure) to facilitate parallel processing on Shared Memory Parallel (SMP) machines. While CAPO currently is integrated seamlessly into CAPTools (developed at the University of Greenwich, now marketed as ParaWise), CAPO was independently developed at Ames Research Center as one of the components for the Legacy Code Modernization (LCM) project. The current version takes serial FORTRAN programs, performs interprocedural data dependence analysis, and generates OpenMP directives. Due to the widely supported OpenMP standard, the generated OpenMP codes have the potential to run on a wide range of SMP machines. CAPO relies on accurate interprocedural data dependence information currently provided by CAPTools. Compiler directives are generated through identification of parallel loops in the outermost level, construction of parallel regions around parallel loops and optimization of parallel regions, and insertion of directives with automatic identification of private, reduction, induction, and shared variables. Attempts also have been made to identify potential pipeline parallelism (implemented with point-to-point synchronization). Although directives are generated automatically, user interaction with the tool is still important for producing good parallel codes. A comprehensive graphical user interface is included for users to interact with the parallelization process.
Jin, Haoqiang↗
GAHLS: an optimized graph analytics based high level synthesis framework
The urgent need for low latency, high-compute and low power on-board intelligence in autonomous systems, cyber-physical systems, robotics, edge computing, evolvable computing, and complex data science calls for determining the optimal amount and type of specialized hardware together with reconfigurability capabilities. With these goals in mind, we propose a novel comprehensive graph analytics based high level synthesis (GAHLS) framework that efficiently analyzes complex high level programs through a combined compiler-based approach and graph theoretic optimization and synthesizes them into message passing domain-specific accelerators. This GAHLS framework first constructs a compiler-assisted dependency graph (CaDG) from low level virtual machine (LLVM) intermediate representation (IR) of high level programs and converts it into a hardware friendly description representation. Next, the GAHLS framework performs a memory design space exploration while account for the identified computational properties from the CaDG and optimizing the system performance for higher bandwidth. The GAHLS framework also performs a robust optimization to identify the CaDG subgraphs with similar computational structures and aggregate them into intelligent processing clusters in order to optimize the usage of underlying hardware resources. Finally, the GAHLS framework synthesizes this compressed specialized CaDG into processing elements while optimizing the system performance and area metrics. Evaluations of the GAHLS framework on several real-life applications (e.g., deep learning, brain machine interfaces) demonstrate that it provides 14.27× performance improvements compared to state-of-the-art approaches such as LegUp 6.2.
97 MATHEMATICS AND COMPUTING↗
PowerMappeR: Power-Optimized Mapping of SNNs onto ReRAM Crossbars coupled via Packet-Switched NoCs
Many recent efforts in developing hardware-accelerated spiking neural networks (SNNs) are characterized by deep co-design between algorithms, architectures, and devices. Architectural advances overcome device constraints by coupling together many small resistive-RAM (ReRAM) crossbars via a network-on-chip (NoC) for neuromorphic component operation. Concurrently, improved SNN training methods increase accuracy and structural sparsity in networks despite growing problem sizes. Finally, compilers leverage these attributes to minimize area and inter-crossbar communication while mapping large SNNs to sophisticated architectures. However, for compiler-driven co-design to realize increasingly complex and profitable optimizations, a compile-time view of power consumption is critical. We present PowerMappeR to express and optimize over mapping-, architecture-, and device-specific power consumption information. By modeling the dynamic power of well-established components, we develop an integer linear programming (ILP)-based, encoding-agnostic, parametric power estimation model. Using this model, we demonstrate practical improvements in area and inter-crossbar communication by 0%–9.5% and 1.4%–5.1%, respectively. We also limit hotspot formation during optimization, achieving comparable or better results in targeted metrics with up to 96.4%–97.1% restriction of hotspot magnitude. Finally, we introduce profile-guided formulations to reduce worst-case and expected-case hotspot magnitude by 40.7%–69.5% and 40.6%–56.3%, respectively. Optimizing worst-case hotspot magnitude incidentally improves expected-case magnitude by 10.85%–33.45%. Reciprocally, optimizing expected-case magnitude incidentally improves worst-case magnitude by 4.33%–39.87%. Validation against hardware simulators confirms that PowerMappeR can decrease dynamic power consumption by 12.6%–27.3%.
Pohl, Devin [ORNL] (ORCID:0009000040149027)↗
Variable C : N : P stoichiometry of dissolved organic matter cycling in the Community Earth System Model
Dissolved organic matter (DOM) plays an important role in the ocean's biological carbon pump by providing an advective/mixing pathway for ~ 20% of export production. DOM is known to have a stoichiometry depleted in nitrogen (N) and phosphorus (P) compared to the particulate organic matter pool, a fact that is often omitted from biogeochemical ocean general circulation models. However the variable C : N : P stoichiometry of DOM becomes important when quantifying carbon export from the upper ocean and linking the nutrient cycles of N and P with that of carbon. Here we utilize recent advances in DOM observational data coverage and offline tracer-modeling techniques to objectively constrain the variable production and remineralization rates of the DOM C : N : P pools in a simple biogeochemical-ocean model of DOM cycling. The optimized DOM cycling parameters are then incorporated within the Biogeochemical Elemental Cycling (BEC) component of the Community Earth System Model (CESM) and validated against the compilation of marine DOM observations. The optimized BEC simulation including variable DOM C : N : P cycling was found to better reproduce the observed DOM spatial gradients than simulations that used the canonical Redfield ratio. Global annual average export of dissolved organic C, N, and P below 100 m was found to be 2.28 Pg C yr -1 (143 Tmol C yr -1 , 16.4 Tmol N yr -1 , and 1 Tmol P yr -1 , respectively, with an average export C : N : P stoichiometry of 225 : 19 : 1 for the semilabile (degradable) DOM pool. Dissolved organic carbon (DOC) export contributed ~ 25% of the combined organic C export to depths greater than 100 m.
Letscher, R. T.↗
KokkACC: Enhancing Kokkos with OpenACC
Template metaprogramming is gaining popularity as a high-level solution for achieving performance portability on heterogeneous computing resources. Kokkos is a representative approach that offers programmers high-level abstractions for generic programming while most of the device-specific code generation and optimizations are delegated to the compiler through template specializations. For this, Kokkos provides a set of device-specific code specializations in multiple back ends, such as CUDA and HIP. Unlike CUDA or HIP, OpenACC is a high-level and directive-based programming model. This descriptive model allows developers to insert hints (pragmas) into their code that help the compiler to parallelize the code. The compiler is responsible for the transformation of the code, which is completely transparent to the programmer. This paper presents an OpenACC back end for Kokkos: KokkACC. As an alternative to Kokkos’s existing device-specific back ends, KokkACC is a multi-architecture back end providing a high-productivity programming environment enabled by OpenACC’s high-level and descriptive programming model. Moreover, we have observed competitive performance; in some cases, KokkACC is faster (up to 9×) than NVIDIA’s CUDA back end and much faster than OpenMP’s GPU offloading back end. This work also includes implementation details and a detailed performance study conducted with a set of mini-benchmarks (AXPY and DOT product) and three mini-apps (LULESH, miniFE and SNAP, a LAMMPS proxy mini-app).
Valero Lara, Pedro↗
Level set topology optimization in Firedrake
Lestofire is an academic software for the automated design of engineering components. It finds the optimal geometry of a system to minimize a given cost function while constrained by a given set of physics equations. Using a level set function to describe the geometry, lestofire uses shapes derivative to evolve the system until its optimal shape. It leverages the compilation of variational forms in Firedrake (https://www.firedrakeproject.org/) and the automated calculation of shape derivatives via pyadjoint (http://www.dolfin-adjoint.or/en/latest/) to allow fast prototyping of different physics, cost functions and design parametrizations.
Salazar De Troya, Miguel↗
Characterizing Computers And Predicting Computing Times
Improved method for evaluation and comparison of computers running same or different FORTRAN programs devised. Enables one to predict time necessary to run given "benchmark" or other standard program on given computer, in scalar mode and without optimization of codes generated by compiler. Such "benchmark" running times are principal measures used to characterize performances of computers; of interest to designers, manufacturers, programmers, and users.
Saavedra-Barrera, Rafael H.↗
DTS: Building custom, intelligent schedulers
DTS is a decision-theoretic scheduler, built on top of a flexible toolkit -- this paper focuses on how the toolkit might be reused in future NASA mission schedulers. The toolkit includes a user-customizable scheduling interface, and a 'Just-For-You' optimization engine. The customizable interface is built on two metaphors: objects and dynamic graphs. Objects help to structure problem specifications and related data, while dynamic graphs simplify the specification of graphical schedule editors (such as Gantt charts). The interface can be used with any 'back-end' scheduler, through dynamically-loaded code, interprocess communication, or a shared database. The 'Just-For-You' optimization engine includes user-specific utility functions, automatically compiled heuristic evaluations, and a postprocessing facility for enforcing scheduling policies. The optimization engine is based on BPS, the Bayesian Problem-Solver (1,2), which introduced a similar approach to solving single-agent and adversarial graph search problems.
Hansson, Othar↗
A domain-specific compiler for a parallel multiresolution adaptive numerical simulation environment
This paper describes the design and implementation of a layered domain-specific compiler to support MADNESS---Multiresolution ADaptive Numerical Environment for Scientific Simulation. MADNESS is a high-level software environment for the solution of integral and differential equations in many dimensions, using adaptive and fast harmonic analysis methods with guaranteed precision. MADNESS uses k-d trees to represent spatial functions and implements operators like addition, multiplication, differentiation, and integration on the numerical representation of functions. The MADNESS runtime system provides global namespace support and a task-based execution model including futures. MADNESS is currently deployed on massively parallel supercomputers and has enabled many science advances. Due to the highly irregular and statically unpredictable structure of the k-d trees representing the spatial functions encountered in MADNESS applications, only purely runtime approaches to optimization have previously been implemented in the MADNESS framework. This paper describes a layered domain-specific compiler developed to address some performance bottlenecks in MADNESS. The newly developed static compile-time optimizations, in conjunction with the MADNESS runtime support, enable significant performance improvement for the MADNESS framework.
Rajbhandari, Samyam↗
SBIR Phase I Final Report, TACO: Distributed and Heterogeneous Sparse Compiler
Tensor algebra is a powerful tool for computing, but writing optimized codes that operate on sparse tensors can be very complex. This project enables a Tensor Algebra Compiler (TACO) that simplifies this task from man-years to man-days and extends TACO to support complex and large distributed systems. This report details the hypotheses, approaches used, and findings in this project.
97 MATHEMATICS AND COMPUTING↗
Compiler generation and autotuning of communication-avoiding operators for geometric multigrid
This paper describes a compiler approach to introducing communication-avoiding optimizations in geometric multigrid (GMG), one of the most popular methods for solving partial differential equations. Communication-avoiding optimizations reduce vertical communication through the memory hierarchy and horizontal communication across processes or threads, usually at the expense of introducing redundant computation. We focus on applying these optimizations to the smooth operator, which successively reduces the error and accounts for the largest fraction of the GMG execution time. Our compiler technology applies both novel and known transformations to derive an implementation comparable to manually-tuned code. To make the approach portable, an underlying autotuning system explores the tradeoff between reduced communication and increased computation, as well as tradeoffs in threading schemes, to automatically identify the best implementation for a particular architecture and at each computation phase. Results show that we are able to quadruple the performance of the smooth operation on the finest grids while attaining performance within 94% of manually-tuned code. Overall we improve the overall multigrid solve time by 2.5× without sacrificing programer productivity.
97 MATHEMATICS AND COMPUTING↗
The ParaScope parallel programming environment
The ParaScope parallel programming environment, developed to support scientific programming of shared-memory multiprocessors, includes a collection of tools that use global program analysis to help users develop and debug parallel programs. This paper focuses on ParaScope's compilation system, its parallel program editor, and its parallel debugging system. The compilation system extends the traditional single-procedure compiler by providing a mechanism for managing the compilation of complete programs. Thus, ParaScope can support both traditional single-procedure optimization and optimization across procedure boundaries. The ParaScope editor brings both compiler analysis and user expertise to bear on program parallelization. It assists the knowledgeable user by displaying and managing analysis and by providing a variety of interactive program transformations that are effective in exposing parallelism. The debugging system detects and reports timing-dependent errors, called data races, in execution of parallel programs. The system combines static analysis, program instrumentation, and run-time reporting to provide a mechanical system for isolating errors in parallel program executions. Finally, we describe a new project to extend ParaScope to support programming in FORTRAN D, a machine-independent parallel programming language intended for use with both distributed-memory and shared-memory parallel computers.
Cooper, Keith D.↗