Floating-point numbers are finite-precision approximations to real numbers and are ubiquitous in computer applications in nearly every field. Selecting the right floating-point representation that balances performance and numerical accuracy is a difficult task – one that has become even more critical as hardware trends toward high-performance, low-precision operations. Although the common wisdom around changing floating-point precision implies that accuracy and performance are inversely correlated, more advanced techniques can often circumvent this tradeoff. Applying complex numerical optimizations to real-world code, however, is an arduous engineering task that requires expertise in numerical analysis and performance engineering, and application-specific numerical context. While there is a plethora of existing tools that partially automate this process, they are limited in the scope of optimization techniques or still require substantial human intervention. We present Poseidon, a modular and extensible framework that fully automates floating-point optimizations for real-world applications within a production compiler. Our key insight is that a small surrogate profile often reveals sufficient numerical context to drive effective rewrites. Poseidon operates as a two-phase compiler: the first compilation instruments the program to capture numerical context; the second compilation consumes profiled data, generates and evaluates candidate rewrites, and solves for optimal performance/accuracy tradeoffs. Poseidon’s interoperability with standard compiler analyses and optimizations grants it analysis and optimization advantages unavailable to existing source- and binary-level approaches. On multiple large-scale applications, Poseidon leads to outsized benefits in performance without substantially changing accuracy, and outsized accuracy benefits without diminishing performance. On a quaternion differentiator, Poseidon enables a 1.46× speedup with a relative error of 10−7. On DOE’s LULESH hydrodynamics application, Poseidon improves program accuracy to exactly match a 512-bit simulation run without substantially reducing performance.

Abstract This paper presents a modular approach to verifying the vector module of PETSc, a widely used library in scientific computing, using the CIVL model checker. Our approach relies on the creation of stub functions, which serve a dual purpose of specifying the intended behavior of individual PETSc functions and providing an abstraction for called functions to allow efficient verification of callers. This facilitates the use of symbolic execution and model checking to establish the correctness of isolated functions. Our work contributes to the ongoing effort to enhance the reliability of high-performance computing libraries and proposes an effective verification strategy for complex scientific software.

The Massachusetts Institute of Technology General Circulation Model (MITgcm) is widely used by the climate science community to simulate planetary atmosphere and ocean circulations. A defining feature of the MITgcm is that it has been developed to be compatible with an algorithmic differentiation (AD) tool, TAF, enabling the generation of tangent-linear and adjoint models. These provide gradient information which enables dynamics-based sensitivity and attribution studies, state and parameter estimation, and rigorous uncertainty quantification. Importantly, gradient information is essential for computing comprehensive sensitivities and performing efficient large-scale data assimilation, ensuring that observations collected from satellites and in-situ measuring instruments can be effectively used to optimize a large uncertain control space. As a result, the MITgcm forms the dynamical core of a key data assimilation product employed by the physical oceanography research community: Estimating the Circulation and Climate of the Ocean (ECCO) state estimate. Although MITgcm and ECCO are used extensively within the research community, the AD tool TAF is proprietary and hence inaccessible to a large proportion of these users. The new version 2 (MITgcm-AD v2) framework introduced here is based on the source-to-source AD tool Tapenade, which has recently been open-sourced. Another feature of Tapenade is that it stores required variables by default (instead of recomputing them) which simplifies the implementation of efficient, AD-compatible code. The framework has been integrated with the MITgcm model’s main branch and is now freely available.

Partial Differential Equations (PDEs) play a ubiquitous role in scientific computing and engineering. While numerical methods make solving PDEs tractable, these numerical solvers encounter several issues, particularly for hyperbolic PDEs. These issues arise from multiple sources including the PDE’s physical model, which can lead to effects like shock wave formation, and the PDE solver’s inherent approximations, which can introduce spurious numerical artifacts. These issues can cause the solver’s program execution to crash (due to overflow) or return results with unacceptable levels of inaccuracy (due to spurious oscillations or dissipation). Moreover, these challenges are compounded by the nonlinear nature of many of these PDEs. In addition, PDE solvers must obey numerical invariants like the CFL condition. Hence there exists a critical need to apply program analysis to PDE solvers to certify such problems do not arise and that invariants are always satisfied. As a solution, we develop Phocus, which is the first abstract interpretation of hyperbolic PDE solvers. Phocus can certify precise bounds on nonlinear PDE solutions and certify key invariants such as the CFL condition and a solution’s total variation bound. Hence Phocus can verify the absence of shock formation, the stability of the solver, and bounds on the amount of spurious numerical effects. To enable effective abstract interpretation of hyperbolic PDE solvers, Phocus uses a novel optimization-based procedure to synthesize precise abstract transformers for multiple finite difference schemes. To evaluate Phocus, we develop a new set of PDE benchmark programs and use them to perform an extensive experimental evaluation which demonstrates Phocus’s significant precision benefits and scalability to several thousand mesh points.

Abstract Many parallel programming models guarantee that if all sequentially consistent (SC) executions of a program are free of data races, then all executions of the program will appear to be sequentially consistent. This greatly simplifies reasoning about the program, but leaves open the question of how to verify that all SC executions are race-free. In this paper, we show that with a few simple modifications, model checking can be an effective tool for verifying race-freedom. We explore this technique on a suite of C programs parallelized with OpenMP.

Alias analysis (AA) is a prerequisite for many compiler optimizations, which are crucial for performance especially for parallel and scientific software. AA is the subject of ongoing research, and compilers can in practice only approximate the alias information of a given program. In this paper we investigate the extent to which performance in high-performance computing (HPC) applications could be improved if better AA were available in LLVM, one of the most widely used compilers today.

This paper presents a novel combination of reverse mode automatic differentiation and formal methods, to enable efficient differentiation of (or backpropagation through) shared-memory parallel loops. Compared to the state of the art, our approach can reduce the need for atomic updates or private data copies during the parallel derivative computation, even in the presence of unstructured or data-dependent data access patterns. This is achieved by gathering information about the memory access patterns from the input program, which is assumed to be correctly parallelized. This information is then used to build a model of assertions in a theorem prover, which can be used to check the safety of shared memory accesses during the parallel derivative loops. We demonstrate this approach on scientific computing benchmarks including a lattice-Boltzmann method (LBM) solver from the Parboil benchmark suite and a Green’s function Monte Carlo (GFMC) kernel from the CORAL benchmark suite.

The accompanying poster to this short paper presents a combination of reverse mode AD and formal methods to enable efficient differentiation of (or backpropagation through) shared-memory parallel code. Compared to the state of the art, our approach can more often avoid the need for atomic updates or private data copies during the parallel derivative computation, even in the presence of unstructured or data-dependent data access patterns. This is achieved by gathering information about the memory access patterns from the input program, which is assumed to be correctly parallelized. This information is then used to build a model of assertions in a theorem prover, which can be used to check the safety of shared memory accesses during the parallel derivative computation.

Derivatives are key to numerous science, engineering, and machine learning applications. While existing tools generate derivatives of programs in a single language, modern parallel applications combine a set of frameworks and languages to leverage available performance and function in an evolving hardware landscape. We propose a scheme for differentiating arbitrary DAG-based parallelism that preserves scalability and efficiency, implemented into the LLVM-based Enzyme automatic differentiation framework. By integrating with a full-fledged compiler backend, Enzyme can differentiate numerous parallel frameworks and directly control code generation. Combined with its ability to differentiate any LLVM-based language, this flexibility permits Enzyme to leverage the compiler tool chain for parallel and differentiation-specitic optimizations. We differentiate nine distinct versions of the LULESH and miniBUDE applications, written in different programming languages (C++, Julia) and parallel frameworks (OpenMP, MPI, RAJA, Julia tasks, MPI.jl), demonstrating similar scalability to the original program. On benchmarks with 64 threads or nodes, we find a differentiation overhead of 3.4–6.8× on C++ and 5.4–12.5× on Julia.

Abstract Fortran is widely used in computational science, engineering, and high performance computing. This paper presents an extension to the CIVL verification framework to check correctness properties of Fortran programs. Unlike previous work that translates Fortran to C, LLVM IR, or other intermediate formats before verification, our work allows CIVL to directly consume Fortran source files. We extended the parsing, translation, and analysis phases to support Fortran-specific features such as array slicing and reshaping, and to find program violations that are specific to Fortran, such as argument aliasing rule violations, invalid use of variable and function attributes, or defects due to Fortran’s unspecified expression evaluation order. We demonstrate the usefulness of our tool on a verification benchmark suite and kernels extracted from a real world application.

This article presents our work toward correct and efficient automatic differentiation of OpenMP parallel worksharing loops in forward and reverse mode. Automatic differentiation is a method to obtain gradients of numerical programs, which are crucial in optimization, uncertainty quantification, and machine learning. The computational cost to compute gradients is a common bottleneck in practice. For applications that are parallelized for multicore CPUs or GPUs using OpenMP, one also wishes to compute the gradients in parallel. We propose a framework to reason about the correctness of the generated derivative code, from which we justify our OpenMP extension to the differentiation model. We implement this model in the automatic differentiation tool Tapenade and present test cases that are differentiated following our extended differentiation procedure. Performance of the generated derivative programs in forward and reverse mode is better than sequential, although our reverse mode often scales worse than the input programs.

We present SPRAY, an open-source header-only C++ library for sparse reductions of arrays. SPRAY is meant for applications in which a large array is collaboratively updated by multiple threads using an associative and commutative operation such as +=. Especially when each thread accesses only parts of the array, SPRAY can perform significantly better than OPENMP's built-in reduction clause or atomic updates, while also using less memory than the former. SPRAY provides both an easy-to-use interface that can serve as a drop-in replacement for OPENMP reductions and a selection of reducer objects that accumulate the final result in different thread-safe ways. We demonstrate SPRAY through multiple test cases including the LULESH shock hydrodynamics code and a transpose-matrix-vector multiplication for sparse matrices stored in CSR format. SPRAY reductions outperform built-in OPENMP reductions consistently, in some cases improving run time and memory overhead by 20X, and even beating domain-specific approaches such as Intel MKL by over 2X in some cases. Furthermore, SPRAY reductions have a minimal impact on the code base, requiring only a few lines of source code changes. Once in place, SPRAY reduction schemes can be switched easily, allowing performance portability and tuning opportunities by separating performance-critical implementation details from application code.

Computing derivatives is key to many algorithms in scientific computing and machine learning such as optimization, uncertainty quantification, and stability analysis. Enzyme is a LLVM compiler plugin that performs reverse-mode automatic differentiation (AD) and thus generates high performance gradients of programs in languages including C/C++, Fortran, Julia, and Rust. Prior to this work, Enzyme and other AD tools were not capable of generating gradients of GPU kernels. Our paper presents a combination of novel techniques that make Enzyme the first fully automatic reversemode AD tool to generate gradients of GPU kernels. Since unlike other tools Enzyme performs automatic differentiation within a general-purpose compiler, we are able to introduce several novel GPU and AD-specific optimizations. To show the generality and efficiency of our approach, we compute gradients of five GPU-based HPC applications, executed on NVIDIA and AMD GPUs. All benchmarks run within an order of magnitude of the original program's execution time. Without GPU and AD-specific optimizations, gradients of GPU kernels either fail to run from a lack of resources or have infeasible overhead. Finally, we demonstrate that increasing the problem size by either increasing the number of threads or increasing the work per thread, does not substantially impact the overhead from differentiation.

Automatic differentiation, back-propagation, differentiable programming and related methods have received widespread attention, due to their ability to compute accurate gradients of numerical programs for optimization, uncertainty quantification, and machine learning. Two strategies are commonly used. The forward mode, which is easy to implement but has an overhead compared to the original program that grows linearly with the number of inputs, and the reverse mode, which can compute gradients for an arbitrary number of program inputs with a constant factor overhead, although the constant can be large, more memory is required, and the implementation is often challenging. Previous literature has shown that the forward mode can be more easily parallelized and vectorized than the reverse mode, but case studies investigating when either mode is the best choice are lacking, especially for modern CPUs and GPUs. In this paper, we demonstrate that the forward mode can outperform the reverse mode for programs with tens or hundreds of directional derivatives, a number that may yet increase if current hardware trends continue.

Stencil computations are a key part of many high-performance computing applications, such as image processing, convolutional neural networks, and finite-difference solvers for partial differential equations. Devito is a framework capable of generating highly optimized code given symbolic equations expressed in Python , specialized in, but not limited to, affine (stencil) codes. The lowering process—from mathematical equations down to C++ code—is performed by the Devito compiler through a series of intermediate representations. Several performance optimizations are introduced, including advanced common sub-expressions elimination, tiling, and parallelization. Some of these are obtained through well-established stencil optimizers, integrated in the backend of the Devito compiler. The architecture of the Devito compiler, as well as the performance optimizations that are applied when generating code, are presented. The effectiveness of such performance optimizations is demonstrated using operators drawn from seismic imaging applications.

Stencil loops are a common motif in computations including convolutional neural networks, structured-mesh solvers for partial differential equations, and image processing. Stencil loops are easy to parallelise, and their fast execution is aided by compilers, libraries, and domain-specific languages. Reverse-mode automatic differentiation, also known as algorithmic differentiation, autodiff, adjoint differentiation, or back-propagation, is sometimes used to obtain gradients of programs that contain stencil loops. Unfortunately, conventional automatic differentiation results in a memory access pattern that is not stencil-like and not easily parallelisable.

Ensemble computations are used to evaluate a function for multiple inputs, for example in uncertainty quantification. Embedded ensemble computations perform several evaluations within the same program, often enabling a reduced overall runtime by exploiting vectorisation and parallelisation opportunities that are not present in individual ensemble members. This is challenging if members take different control flow paths. We present a source-to-source transformation that turns a given C program into an embedded ensemble program that computes members in a single-instruction-multiple-data fashion using OpenMP SIMD pragmas. We use techniques from whole-function vectorisation, achieving effective vectorisation for moderate amounts of branch divergence, particularly on processors with masked instructions such as recent Xeon Phi or Skylake processors with AVX-512.

Algorithmic differentiation (AD) by source-transformation is an established method for computing derivatives of computational algorithms. Static dataflow analysis is commonly used by AD tools to determine the set of active variables, that is, variables that are influenced by the program input in a differentiable way and have a differentiable influence on the program output. In this work, a context-sensitive static analysis combined with procedure cloning is used to generate specialised versions of differentiated procedures for each call site. This enables better detection and elimination of unused computations and memory storage, resulting in performance improvements of the generated code, in both forward- and reverse-mode AD. The implications of this multi-activity AD approach on the static analysis of an AD tool is shown using dataflow equations. The worst-case cost of multi-activity AD on the differentiation process is analysed and practical remedies to avoid running into this worst case are presented. The method was implemented in the AD tool Tapenade, and we present its application to a 3D unstructured compressible flow solver, for which we generate an adjoint solver that performs significantly faster when multi-activity AD is used.