In advancing parallel programming, particularly with OpenMP, the shift towards NLP-based methods marks a significant innovation beyond traditional S2S tools like Autopar and Cetus. These NLP approaches train on extensive datasets of examples to efficiently generate optimized parallel code, streamlining the development process. This method's strength lies in its ability to swiftly produce parallelized code that runs efficiently. However, this reliance on NLP models, without direct code analysis, can introduce inaccuracies, as these models might not fully grasp the nuanced semantics of the code they parallelize. We build OMP-Engineer, which balances the efficiency and scalability of NLP models with the accuracy and reliability of traditional methods, aiming to enhance the performance of automating parallelization while navigating its inherent challenges.
Despite the various research initiatives and proposed programming models, efficient solutions for parallel programming in HPC clusters still rely on a complex combination of different programming models (e.g., OpenMP and MPI), languages (e.g., C++ and CUDA), and specialized runtimes (e.g., Charm++ and Legion). On the other hand, task parallelism has shown to be an efficient and seamless programming model for clusters. This paper introduces OpenMP Cluster (OMPC), a task-parallel model that extends OpenMP for cluster programming. OMPC leverages OpenMP's offloading standard to distribute annotate…
Multicore has emerged as a typical architecture model since its advent and stands now as a standard. The trend is to increase the number of cores and improve the performance of the memory system. Providing an efficient multicore implementation for a important algorithmic kernel is a genuine contribution. From a methodology standpoint, this should be done at the level of the underlying paradigm if any. In this paper, we study the cases of {\em dynamic programming} and {\em greedy algorithms}, which are two major algorithmic paradigms. We exclusively consider directives-based loop parallelizatio…
Parallelization schemes are essential in order to exploit the full benefits of multi-core architectures. In said architectures, the most comprehensive parallelization API is OpenMP. However, the introduction of correct and optimal OpenMP parallelization to applications is not always a simple task, due to common parallel management pitfalls, architecture heterogeneity and the current necessity for human expertise in order to comprehend many fine details and abstract correlations. To ease this process, many automatic parallelization compilers were created over the last decade. Harel et al. [2020…
OpenMP provides programmers with directives to parallelize DOALL loops such as parallel for and, more recently, taskloop for task-based parallelism. On the other hand, when it is possible to prove that a loop is DOACROSS, programmers can try to parallelize it through parallel for and to use the OpenMP ordered directive to mark the region of the loop that has to be executed sequentially. However, when neither of the previous two cases can be proven, programmers have to be conservative and assume that the loop is DOACROSS (actually may DOACROSS). Previous work proposed speculative support for ta…
Advising OpenMP Parallelization via a Graph-Based Approach with Transformers
Ben-Gurion University of the Negev（内盖夫本-古里安大学）；Israel Atomic Energy Commission（以色列原子能委员会）；Intel Labs（英特尔实验室）；Nuclear Research Center – Negev（内盖夫核研究中心）；Technion – Israel Institute of Technology（以色列理工学院）
We present our experience with the modernization on the GR-MHD code BHAC, aimed at improving its novel hybrid (MPI+OpenMP) parallelization scheme. In doing so, we showcase the use of performance profiling tools usable on x86 (Intel-based) architectures. Our performance characterization and threading analysis provided guidance in improving the concurrency and thus the efficiency of the OpenMP parallel regions. We assess scaling and communication patterns in order to identify and alleviate MPI bottlenecks, with both runtime switches and precise code interventions. The performance of optimized ve…
Currently, multi/many-core CPUs are considered standard in most types of computers including, mobile phones, PCs or supercomputers. However, the parallelization of applications as well as refactoring/design of applications for efficient hardware usage remains restricted to experts who have advanced technical knowledge and who can invest time tuning their software. In this context, the compilation community has proposed different methods for automatic parallelization, but their focus is traditionally on loops and nested loops with the support of polyhedral techniques. In this study, we propose …
Ben-Gurion University of the Negev（内盖夫本-古里安大学）；Nuclear Research Center – Negev（内盖夫核研究中心）；Technion – Israel Institute of Technology（以色列理工学院）
In past years, the world has switched to many-core and multi-core shared memory architectures. As a result, there is a growing need to utilize these architectures by introducing shared memory parallelization schemes to software applications. OpenMP is the most comprehensive API that implements such schemes, characterized by a readable interface. Nevertheless, introducing OpenMP into code is challenging due to pervasive pitfalls in management of parallel shared memory. To facilitate the performance of this task, many source-to-source (S2S) compilers have been created over the years, tasked with…
In this paper, a work-optimal parallelization of Kostelec and Rockmore's well-known fast Fourier transform and its inverse on the three-dimensional rotation group SO(3) is designed, implemented, and tested. To this end, the sequential algorithms are reviewed briefly first. In the subsequent design and implementation of the parallel algorithms, we use the well-known Forster (PCAM) method and the OpenMP standard. The parallelization itself is based on symmetries of the underlying basis functions and a geometric approach in which the resulting index range is transformed in such a way that distinc…
Task-based programming models like OmpSs-2 and OpenMP provide a flexible data-flow execution model to exploit dynamic, irregular and nested parallelism. Providing an efficient implementation that scales well with small granularity tasks remains a challenge, and bottlenecks can manifest in several runtime components. In this paper, we analyze the limiting factors in the scalability of a task-based runtime system and propose individual solutions for each of the challenges, including a wait-free dependency system and a novel scalable scheduler design based on delegation. We evaluate how the optim…
We investigate a parallelization strategy for dense matrix factorization (DMF) algorithms, using OpenMP, that departs from the legacy (or conventional) solution, which simply extracts concurrency from a multithreaded version of BLAS. This approach is also different from the more sophisticated runtime-assisted implementations, which decompose the operation into tasks and identify dependencies via directives and runtime support. Instead, our strategy attains high performance by explicitly embedding a static look-ahead technique into the DMF code, in order to overcome the performance bottleneck o…
We use OpenMP to target hardware accelerators (GPUs) on Summit, a newly deployed supercomputer at the Oak Ridge Leadership Computing Facility (OLCF), demonstrating simplified access to GPU devices for users of our astrophysics code GenASiS and useful speedup on a sample fluid dynamics problem. We modify our workhorse class for data storage to include members and methods that significantly streamline the persistent allocation of and association to GPU memory. Users offload computational kernels with OpenMP target directives that are rather similar to constructs already familiar from multi-core …
Large scale plasma instabilities inside a tokamak can be influenced by the currents flowing in the conducting vessel wall. This involves non linear plasma dynamics and its interaction with the wall current. In order to study this problem the code that solves the magneto-hydrodynamic (MHD) equations, called JOREK [Huysmans G.T.A. and Czarny O. NF 47, 659 (2007); Czarny O. and Huysmans G. JCP 227, 7423 (2008)], was coupled [Hoelzl M., et al. Journal of Physics: Conference Series, 401, 012010 (2012)] with the model for the vacuum region and the resistive conducting structure named STARWALL [Merke…
Approximation via sampling is a widespread technique whenever exact solutions are too expensive. In this paper, we present techniques for an efficient parallelization of adaptive (a. k. a. progressive) sampling algorithms on multi-threaded shared-memory machines. Our basic algorithmic technique requires no synchronization except for atomic load-acquire and store-release operations. It does, however, require O(n) memory per thread, where n is the size of the sampling state. We present variants of the algorithm that either reduce this memory consumption to O(1) or ensure that deterministic resul…
This article describes a method for creating applications for cluster computing systems using the parallel BSF skeleton based on the original BSF (Bulk Synchronous Farm) model of parallel computations developed by the author earlier. This model uses the master/slave paradigm. The main advantage of the BSF model is that it allows to estimate the scalability of a parallel algorithm before its implementation. Another important feature of the BSF model is the representation of problem data in the form of lists that greatly simplifies the logic of building applications. The BSF skeleton is designed…
The main computing tasks of a finite element code(FE) for solving partial differential equations (PDE's) are the algebraic system assembly and the iterative solver. This work focuses on the first task, in the context of a hybrid MPI+X paradigm. Although we will describe algorithms in the FE context, a similar strategy can be straightforwardly applied to other discretization methods, like the finite volume method. The matrix assembly consists of a loop over the elements of the MPI partition to compute element matrices and right-hand sides and their assemblies in the local system to each MPI par…
The Gaia Astrometric Verification Unit-Global Sphere Reconstruction (AVU-GSR) Parallel Solver aims to find the astrometric parameters for $\sim$10$^8$ stars in the Milky Way, the attitude and the instrumental specifications of the Gaia satellite, and the global parameter $γ$ of the post Newtonian formalism. The code iteratively solves a system of linear equations, $\mathbf{A} \times \vec{x} = \vec{b}$, where the coefficient matrix $\mathbf{A}$ is large ($\sim$$10^{11} \times 10^8$ elements) and sparse. To solve this system of equations, the code exploits a hybrid implementation of the iterativ…
