Vassil Vassilev

Just as data sets from next-generation experiments grow, processing requirements for physics analysis become more computationally demanding, necessitating performance optimizations for RooFit. One possibility to speed-up minimization and add stability is the use of Automatic Differentiation (AD). Unlike for numerical differentiation, the computation cost scales linearly with the number of parameters, making AD particularly appealing for statistical models with many parameters. In this paper, we report on one possible way to implement AD in RooFit. Our approach is to add a facility to generate C++ code for a full RooFit model automatically. Unlike the original RooFit model, this generated code is free of virtual function calls and other RooFit-specific overhead. In particular, this code is then used to produce the gradient automatically with Clad. Clad is a source transformation AD tool implemented as a plugin to the clang compiler, which automatically generates the derivative code for input C++ functions. We show results demonstrating the improvements observed when applying this code generation strategy to HistFactory and other commonly used RooFit models. HistFactory is the subcomponent of RooFit that implements binned likelihood models with probability densities based on histogram templates. These models frequently have a very large number of free parameters and are thus an interesting first target for AD support in RooFit.

The simplicity of Python and the power of C++ force stark choices on a scientific software stack. There have been multiple developments to mitigate language boundaries by implementing language bindings, but the impedance mismatch between the static nature of C++ and the dynamic one of Python hinders their implementation; examples include the use of user-defined Python types with templated C++ and advanced memory management. The development of the C++ interpreter Cling has changed the way we can think of language bindings as it provides an incremental compilation infrastructure available at runtime. That is, Python can interrogate C++ on demand, and bindings can be lazily constructed at runtime. This automatic binding provision requires no direct support from library authors and offers better performance than alternative solutions, such as PyBind11. ROOT pioneered this approach with PyROOT, which was later enhanced with its successor, cppyy. However, until now, cppyy relied on the reflection layer of ROOT, which is limited in terms of provided features and performance. This paper presents the next step for language interoperability with cppyy, enabling research into uniform cross-language execution environments and boosting optimization opportunities across language boundaries. We illustrate the use of advanced C++ in Numba-accelerated Python through cppyy. We outline a path forward for re-engineering parts of cppyy to use upstream LLVM components to improve performance and sustainability. We demonstrate cppyy purely based on a C++ reflection library, InterOp , which offers interoperability primitives based on Cling and Clang-Repl.

The Cling C++ interpreter has transformed language bindings by enabling incremental compilation at runtime. This allows Python to interact with C++ on demand and lazily construct bindings between the two. The emergence of Clang-Repl, within the LLVM compiler framework, as a potential alternative to Cling highlights the need for a unified framework for interactive C++ technologies. We present CppInterOp, a C++ interoperability library that leverages Cling and LLVM’s Clang-Repl to provide a minimalist and backward-compatible API facilitating seamless language interoperability. This provides downstream interactive C++ tools with the compiler as a service by embedding Clang and LLVM as libraries in their codebases. By enabling dynamic Python interactions with static C++ codebases, CppInterOp enhances computational efficiency and rapid development in high-energy physics. The library offers reflection and ondemand JIT compilation APIs enabling cppyy, an automatic, run-time, Python- C++ bindings generator. We also demonstrate CppInterOp’s utility in diverse computing environments through its adoption as the runtime engine for xeuscpp, a Jupyter kernel designed for C++. CppInterOp is a general-purpose library inspired by the developments in the ROOT framework, which pushed the frontiers of interactive C++. It aims to extend this approach and serve as an integral component of ROOT, enhancing performance and resilience. This article introduces CppInterOp to the HEP community and showcases how it optimizes cross-language execution and computational tasks in high-energy physics, making it a valuable tool for researchers and developers.

ROOT is an open source framework, freely available on GitHub, at the heart of data acquisition, processing and analysis of HE(N)P experiments, and beyond. It is developed collaboratively: contributions are not authored only by ROOT team members, but also by the user community at large: developers and scientists from universities, labs as well as the private sector. More than 1500 GitHub Pull Requests are merged on average per year. It is in this context that code integration acquires a primary role. The review of code contributions isn’t enough: not only they need to be thoroughly reviewed, they also need to be thoroughly tested through a powerful CI infrastructure on several different platforms to comply with the high code quality standards of the project. Since the end of 2023, ROOT moved its continuous integration system from Jenkins to GitHub Actions. In this contribution, we characterise the transition to the GitHub CI, focussing on our strategy, its implementation and the lessons learned, as well as the advantages the new system offers with respect to the previous one. Particular emphasis will be given to the evaluation of the cost-benefit ratio for Jenkins and GitHub Actions for the ROOT project. We also describe how we manage to run in less than one hour thousands of unit, integration, functional and end-to-end tests on different flavours of Windows, four versions of macOS, as well as about ten of the most used Linux distributions, taking advantage of the CERN computing infrastructure.

Among the well-known methods to approximate derivatives of expectancies computed by Monte-Carlo simulations, averages of pathwise derivatives are often the easiest one to apply. Computing them via algorithmic differentiation typically does not require major manual analysis and rewriting of the code, even for very complex programs like simulations of particle-detector interactions in high-energy physics. However, the pathwise derivative estimator can be biased if there are discontinuities in the program, which may diminish its value for applications. This work integrates algorithmic differentiation into the electromagnetic shower simulation code HepEmShow based on G4HepEm, allowing us to study how well pathwise derivatives approximate derivatives of energy depositions in a sampling calorimeter with respect to parameters of the beam and geometry. We found that when multiple scattering is disabled in the simulation, means of pathwise derivatives converge quickly to their expected values, and these are close to the actual derivatives of the energy deposition. Additionally, we demonstrate the applicability of this novel gradient estimator for stochastic gradient-based optimization in a model example.

As we reach the limit of Moore's Law, researchers are exploring different paradigms to achieve unprecedented performance. Approximate Computing (AC), which relies on the ability of applications to tolerate some error in the results to trade-off accuracy for performance, has shown significant promise. Despite the success of AC in domains such as Machine Learning, its acceptance in High-Performance Computing (HPC) is limited due to stringent requirements for accuracy. We need tools and techniques to identify regions of code that are amenable to approximations and their impact on the application output quality to guide developers to employ selective approximation. To this end, we propose CHEF-FP, a flexible, scalable, and easy-to-use source-code transformation tool based on Automatic Differentiation (AD) for analyzing approximation errors in HPC applications. CHEF-FP uses Clad, an efficient AD tool built as a plugin to the Clang compiler and based on the LLVM compiler infrastructure, as a backend and utilizes its AD abilities to evaluate approximation errors in C++ code. CHEF-FP works at the source by injecting error estimation code into the generated adjoints. This enables the error-estimation code to undergo compiler optimizations resulting in improved analysis time and reduced memory usage. We also provide theoretical and architectural augmentations to source code transformation-based AD tools to perform FP error analysis. This paper primarily focuses on analyzing errors introduced by mixed-precision AC techniques. We also show the applicability of our tool in estimating other kinds of errors by evaluating our tool on codes that use approximate functions. Moreover, we demonstrate the speedups CHEF-FP achieved during analysis time compared to the existing state-of-the-art tool due to its ability to generate and insert approximation error estimate code directly into the derivative source.

The simplicity of Python and the power of C++ force stark choices on a scientific software stack. There have been multiple developments to mitigate language boundaries by implementing language bindings, but the impedance mismatch between the static nature of C++ and the dynamic one of Python hinders their implementation; examples include the use of user-defined Python types with templated C++ and advanced memory management. The development of the C++ interpreter Cling has changed the way we can think of language bindings as it provides an incremental compilation infrastructure available at runtime. That is, Python can interrogate C++ on demand, and bindings can be lazily constructed at runtime. This automatic binding provision requires no direct support from library authors and offers better performance than alternative solutions, such as PyBind11. ROOT pioneered this approach with PyROOT, which was later enhanced with its successor, cppyy. However, until now, cppyy relied on the reflection layer of ROOT, which is limited in terms of provided features and performance. This paper presents the next step for language interoperability with cppyy, enabling research into uniform cross-language execution environments and boosting optimization opportunities across language boundaries. We illustrate the use of advanced C++ in Numba-accelerated Python through cppyy. We outline a path forward for re-engineering parts of cppyy to use upstream LLVM components to improve performance and sustainability. We demonstrate cppyy purely based on a C++ reflection library, InterOp, which offers interoperability primitives based on Cling and Clang-Repl.

RooFit is a toolkit for statistical modeling and fitting used by most experiments in particle physics. Just as data sets from next-generation experiments grow, processing requirements for physics analysis become more computationally demanding, necessitating performance optimizations for RooFit. One possibility to speed-up minimization and add stability is the use of Automatic Differentiation (AD). Unlike for numerical differentiation, the computation cost scales linearly with the number of parameters, making AD particularly appealing for statistical models with many parameters. In this paper, we report on one possible way to implement AD in RooFit. Our approach is to add a facility to generate C++ code for a full RooFit model automatically. Unlike the original RooFit model, this generated code is free of virtual function calls and other RooFit-specific overhead. In particular, this code is then used to produce the gradient automatically with Clad. Clad is a source transformation AD tool implemented as a plugin to the clang compiler, which automatically generates the derivative code for input C++ functions. We show results demonstrating the improvements observed when applying this code generation strategy to HistFactory and other commonly used RooFit models. HistFactory is the subcomponent of RooFit that implements binned likelihood models with probability densities based on histogram templates. These models frequently have a very large number of free parameters and are thus an interesting first target for AD support in RooFit.

Automatic Differentiation (AD) is instrumental for science and industry. It is a tool to evaluate the derivative of a function specified through a computer program. The range of AD application domain spans from Machine Learning to Robotics to High Energy Physics. Computing gradients with the help of AD is guaranteed to be more precise than the numerical alternative and have a low, constant factor more arithmetical operations compared to the original function. Moreover, AD applications to domain problems typically are computationally bound. They are often limited by the computational requirements of high-dimensional parameters and thus can benefit from parallel implementations on graphics processing units (GPUs). Clad aims to enable differential analysis for C/C++ and CUDA and is a compiler-assisted AD tool available both as a compiler extension and in ROOT. Moreover, Clad works as a plugin extending the Clang compiler; as a plugin extending the interactive interpreter Cling; and as a Jupyter kernel extension based on xeus-cling. We demonstrate the advantages of parallel gradient computations on GPUs with Clad. We explain how to bring forth a new layer of optimization and a proportional speed up by extending Clad to support CUDA. The gradients of well-behaved C++ functions can be automatically executed on a GPU. The library can be easily integrated into existing frameworks or used interactively. Furthermore, we demonstrate the achieved application performance improvements, including (~10x) in ROOT histogram fitting and corresponding performance gains from offloading to GPUs.