Supercomputer Architectures Research Articles

An Efficient RI-MP2 Algorithm for Distributed Many-GPU Architectures.

Second-order Møller-Plesset perturbation theory (MP2) using the Resolution of the Identity approximation (RI-MP2) is a widely used method for computing molecular energies beyond the Hartree-Fock mean-field approximation. However, its high computational cost and lack of efficient algorithms for modern supercomputing architectures limit its applicability to large molecules. In this paper, we present the first distributed-memory many-GPU RI-MP2 algorithm explicitly designed to utilize hundreds of GPU accelerators for every step of the computation. Our novel algorithm achieves near-peak performance on GPU-based supercomputers through the development of a distributed memory algorithm for forming RI-MP2 intermediate tensors with zero internode communication, except for a single asynchronous broadcast, and a distributed memory algorithm for the energy reduction step, capable of sustaining near-peak performance on clusters with several hundred GPUs. Comparative analysis shows our implementation outperforms state-of-the-art quantum chemistry software by over 3.5 times in speed while achieving an 8-fold reduction in computational power consumption. Benchmarking on the Perlmutter supercomputer, our algorithm achieves 11.8 PFLOP/s (83% of peak performance) performing and the RI-MP2 energy calculation on a 314-water cluster with 7850 primary and 30,144 auxiliary basis functions in 4 min on 180 nodes and 720 A100 GPUs. This performance represents a substantial improvement over traditional CPU-based methods, demonstrating significant time-to-solution and power consumption benefits of leveraging modern GPU-accelerated computing environments for quantum chemistry calculations.

Effect of molybdenum addition on precipitate coarsening kinetics in Inconel 740H: A phase-field study

Inconel 740H (IN740H) has emerged as an important candidate for the advanced ultra-supercritical (AUSC) steam turbines due to its superior microstructural stability and creep resistance under service conditions. In the present work, we have reparameterized a multicomponent Ni-based superalloy (IN740H) into an equivalent ternary Ni-Al-Mo superalloy, based on the partitioning coefficients of the elements, using the thermodynamic database (CALPHAD). We use this thermodynamic description to employ a quantitative phase-field model to assess the long-term stability of γ′ precipitates in IN740H utilizing GPU-based supercomputing architecture. The assessment helps us to enhance our understanding of the effect of the atomic diffusivity of Mo on coarsening kinetics of the γ′-precipitates in equivalent ternary Ni-Al-Mo superalloy. Investigation reveals that our phase-field model can accurately predict the experimentally observed coarsening kinetics in IN740H.

URANOS-2.0: Improved performance, enhanced portability, and model extension towards exascale computing of high-speed engineering flows

We present URANOS-2.0, the second major release of our massively parallel, GPU-accelerated solver for compressible wall flow applications. This latest version represents a significant leap forward in our initial tool, which was launched in 2023 (De Vanna et al. [1]), and has been specifically optimized to take full advantage of the opportunities offered by the cutting-edge pre-exascale architectures available within the EuroHPC JU. In particular, URANOS-2.0 emphasizes portability and compatibility improvements with the two top-ranked supercomputing architectures in Europe: LUMI and Leonardo. These systems utilize different GPU architectures, AMD and NVIDIA, respectively, which necessitates extensive efforts to ensure seamless usability across their distinct structures. In pursuit of this objective, the current release adheres to the OpenACC standard. This choice not only facilitates efficient utilization of the full potential inherent in these extensive GPU-based architectures but also upholds the principles of vendor neutrality, a distinctive characteristic of URANOS solvers in the CFD solvers' panorama. However, the URANOS-2.0 version goes beyond the goals of improving usability and portability; it introduces performance enhancements and restructures the most demanding computational kernels. This translates into a 2× speedup over the same architecture. In addition to its enhanced single-GPU performance, the present solver release demonstrates very good scalability in multi-GPU environments. URANOS-2.0, in fact, achieves strong scaling efficiencies of over 80% across 64 compute nodes (256 GPUs) for both LUMI and Leonardo. Furthermore, its weak scaling efficiencies reach approximately 95% and 90% on LUMI and Leonardo, respectively, when up to 256 nodes (1024 GPUs) are considered. These significant performance advancements position URANOS-2.0 as a state-of-the-art supercomputing platform tailored for compressible wall turbulence applications, establishing the solver as an integrated tool for various aerospace and energy engineering applications, which can span from direct numerical simulations, wall-resolved large eddy simulations, up to most recent wall-modeled large eddy simulations. Program summaryProgram title: Unsteady Robust All-around Navier-StOkes Solver (URANOS)CPC Library link to program files:https://doi.org/10.17632/pw5hshn9k6.2Developer's repository link:https://github.com/uranos-gpu/uranos-gpu, https://github.com/uranos-gpu/uranos-gpu/tree/v2.0Licensing provisions: BSD License 2.0Programming language: Modern Fortran, OpenACC, MPINature of problem: Solving the compressible Navier-Stokes equations in a three-dimensional Cartesian framework.Solution method: Convective terms are treated with high-resolution shock-capturing schemes. The system dynamics is advanced in time with a three-stage Runge-Kutta method. Parallelization adopts MPI+OpenACC.

Born-Oppenheimer Molecular Dynamics with a Linear Combination of Atomic Orbitals and Hybrid Functionals for Condensed Matter Simulations Made Possible. Theory and Performance for the Microcanonical and Canonical Ensembles.

The implementation of an original Born-Oppenheimer molecular dynamics module is presented, which is able to perform simulations of large and complex condensed phase systems for sufficiently long time scales at the level of density functional theory with hybrid functionals, in the microcanonical (NVE) and canonical (NVT) ensembles. The algorithm is fully integrated in the Crystal code, a program for quantum mechanical simulations of materials, whose peculiarity stems from the use of atom-centered basis functions within a linear combination of atomic orbitals to describe the wave function. The corresponding efficiency in the evaluation of the exact Fock exchange series has led to the implementation of a rich variety of hybrid density functionals at a low computational cost. In addition, the molecular dynamics implementation benefits also from the effective MPI parallelization of the code, suited to exploit high-performance computing resources available on current generation supercomputer architectures. Furthermore, the information contained in the trajectory of the dynamics is extracted through a series of postprocessing algorithms that provide the radial distribution function, the diffusion coefficient and the vibrational density of states. In this work, we present a detailed description of the theoretical framework and the algorithmic implementation, followed by a critical evaluation of the accuracy and parallel performance (e.g., strong and weak scaling) of this approach, when ice and liquid water simulations are performed in the microcanonical and canonical ensembles.

Accuracy of the explicit energy-conserving particle-in-cell method for under-resolved simulations of capacitively coupled plasma discharges

The traditional explicit electrostatic momentum-conserving particle-in-cell algorithm requires strict resolution of the electron Debye length to deliver numerical stability and accuracy. The explicit electrostatic energy-conserving particle-in-cell algorithm alleviates this constraint with minimal modification to the traditional algorithm, retaining its simplicity, ease of parallelization, and acceleration on modern supercomputing architectures. In this article, we apply the algorithm to model a one-dimensional radio frequency capacitively coupled plasma discharge relevant to industrial applications. The energy-conserving approach closely matches the results from the momentum-conserving algorithm and retains accuracy even for cell sizes up to 8 times the electron Debye length. For even larger cells, the algorithm loses accuracy due to poor resolution of steep gradients within the radio frequency sheath. Accuracy can be recovered by adopting a non-uniform grid, which resolves the sheath and allows for cell sizes up to 32 times the electron Debye length in the quasi-neutral bulk of the discharge. The effect is an up to 8 times reduction in the number of required simulation cells, an improvement that can compound in higher-dimensional simulations. We therefore consider the explicit energy-conserving algorithm as a promising approach to significantly reduce the computational cost of full-scale device simulations and a pathway to delivering kinetic simulation capabilities of use to industry.

Earth system modeling on modular supercomputing architecture: coupled atmosphere–ocean simulations with ICON 2.6.6-rc

Abstract. The confrontation of complex Earth system model (ESM) codes with novel supercomputing architectures poses challenges to efficient modeling and job submission strategies. The modular setup of these models naturally fits a modular supercomputing architecture (MSA), which tightly integrates heterogeneous hardware resources into a larger and more flexible high-performance computing (HPC) system. While parts of the ESM codes can easily take advantage of the increased parallelism and communication capabilities of modern GPUs, others lag behind due to the long development cycles or are better suited to run on classical CPUs due to their communication and memory usage patterns. To better cope with these imbalances between the development of the model components, we performed benchmark campaigns on the Jülich Wizard for European Leadership Science (JUWELS) modular HPC system. We enabled the weather and climate model Icosahedral Nonhydrostatic (ICON) to run in a coupled atmosphere–ocean setup, where the ocean and the model I/O is running on the CPU Cluster, while the atmosphere is simulated simultaneously on the GPUs of JUWELS Booster (ICON-MSA). Both atmosphere and ocean are running globally with a resolution of 5 km. In our test case, an optimal configuration in terms of model performance (core hours per simulation day) was found for the combination of 84 GPU nodes on the JUWELS Booster module to simulate the atmosphere and 80 CPU nodes on the JUWELS Cluster module, of which 63 nodes were used for the ocean simulation and the remaining 17 nodes were reserved for I/O. With this configuration the waiting times of the coupler were minimized. Compared to a simulation performed on CPUs only, the MSA approach reduces energy consumption by 45 % with comparable runtimes. ICON-MSA is able to scale up to a significant portion of the JUWELS system, making best use of the available computing resources. A maximum throughput of 170 simulation days per day (SDPD) was achieved when running ICON on 335 JUWELS Booster nodes and 268 Cluster nodes.

SOD2D: A GPU-enabled Spectral Finite Elements Method for compressible scale-resolving simulations

As new supercomputer architectures become more heavily focused on using hardware accelerators, in particular general-purpose graphical processors, it is therefore relevant that algorithms for computational fluid dynamics, especially those targeting scale-resolving simulations, be designed in such a way as to make efficient use of such hardware.In this paper, we propose one such hardware-accelerated Continuous Galerkin Finite Elements model, aimed at handling simulations of turbulent compressible flows over complex geometries. As this model is intended for use in Large-Eddy and Direct Numerical simulations, it is necessary that the resulting scheme introduces only small amounts of artificial (numerical) diffusion for stabilization purposes. We achieve this through a combination of Spectral Finite Elements, operator splittings on the convective term, and the use of the Entropy Viscosity stabilization model adapted to spectral elements scheme.The paper will present the resulting algorithm, how it is made to work efficiently on the accelerators, and results obtained that demonstrate its high-accuracy capabilities.

FerroX: A GPU-accelerated, 3D phase-field simulation framework for modeling ferroelectric devices

We present a massively parallel, 3D phase-field simulation framework for modeling ferroelectric materials based scalable logic devices. This code package, FerroX, self-consistently solves the time-dependent Ginzburg Landau (TDGL) equation for ferroelectric polarization, Poisson's equation for electric potential, and charge equation for carrier densities in semiconductor regions. The algorithm is implemented using the AMReX software framework [1], which provides effective scalability on manycore and GPU-based supercomputing architectures. We demonstrate the performance of the algorithm with excellent scaling results on NERSC multicore and GPU systems, with a significant (15×) speedup on the GPU using a node-by-node comparison. We further demonstrate the applicability of the code in simulations of ferroelectric domain-wall induced negative capacitance (NC) effect in Metal-Ferroelectric-Insulator-Metal (MFIM) and Metal-Ferroelectric-Insulator-Semiconductor-Metal (MFISM) devices. The charge (Q) v.s. voltage (V) responses for these 3D structures clearly indicate stabilized negative capacitance with multidomain formation, which is corroborated by amplification of the voltage at the interface between the ferroelectric and dielectric layers.

Characterizing the performance of node-aware strategies for irregular point-to-point communication on heterogeneous architectures

Supercomputer architectures are trending toward higher computational throughput due to the inclusion of heterogeneous compute nodes. These multi-GPU nodes increase on-node computational efficiency, while also increasing the amount of data to be communicated and the number of potential data flow paths. In this work, we characterize the performance of irregular point-to-point communication with MPI on heterogeneous compute environments through performance modeling, demonstrating the limitations of standard communication strategies for both device-aware and staging-through-host communication techniques. Presented models suggest staging communicated data through host processes then using node-aware communication strategies for high inter-node message counts. Notably, the models also predict that node-aware communication utilizing all available CPU cores to communicate inter-node data leads to the most performant strategy when communicating with a high number of nodes. Model validation is provided via a case study of irregular point-to-point communication patterns in distributed sparse matrix–vector products. Importantly, we include a discussion on the implications model predictions have on communication strategy design for emerging supercomputer architectures.

Specialized Neural Network Hardware Accelerators

It seems that by now enough samples of specialized neural network microprocessor crystals and systems based on them have already been created to indicate the trends of their development, and most importantly, their place in the overall development of supercomputer architectures and technologies. The use of low-bit representations of numbers, such as FP8, INT8, BF16, acceptable in neural network computing, allows, on the one hand, to achieve the performance of the 2015 FP8 TFLOPS, 1008 BF16 TFLOPS crystal, and, on the other hand, to reduce the energy consumption of the multiplication operation. Low bit depth caused attention to rounding errors. In a number of crystals, the set of rounding modes has been expanded in comparison with the generally accepted standard and the possibility of programmatically setting the rounding mode has been introduced. In addition, the validity of the creation of specialized neuroprocessor crystals is due to the use of structural programming elements, in which a computer is programmatically formed for an executable algorithm. Therefore, along with reduction the bit depth and support for processing sparse neural networks, in computing systems created on the basis of SambaNova SN30 RDU, Graphcore Colossus MK2 IPU, Untether AI Boqueria, AWS Trainium1, Tesla Dojo D1, there is the possibility, to some extent, of implementing structural programming of calculations. The sparsity of the processed data caused the abandonment of cache memory and the use of on-chip large scratchpad memory with increased bandwidth for data delivery between memory and arithmetic logic devices, as well as between memory and on-chip and inter-chip communication fabric. Therefore, we can talk about a different hierarchical structure of memory compared to the traditional one using cache memory. Thus, specialization in neural network algorithms has led to the emergence of massively parallel systems architectures for processing low-bit data formats with poor temporal and spatial localization of memory requests.

Insights from high-fidelity modeling of industrial rotary bell atomization

The global automotive industry sprayed over 2.6 billion liters of paint in 2018, much of which through electrostatic rotary bell atomization, a highly complex process involving the fluid mechanics of rapidly rotating thin films tearing apart into micrometer-thin filaments and droplets. Coating operations account for 65% of the energy usage in a typical automotive assembly plant, representing 10,000s of gigawatt-hours each year in the United States alone. Optimization of these processes would allow for improved robustness, reduced material waste, increased throughput, and significantly reduced energy usage. Here, we introduce a high-fidelity mathematical and algorithmic framework to analyze rotary bell atomization dynamics at industrially relevant conditions. Our approach couples laboratory experiment with the development of robust non-Newtonian fluid models; devises high-order accurate numerical methods to compute the coupled bell, paint, and gas dynamics; and efficiently exploits high-performance supercomputing architectures. These advances have yielded insight into key dynamics, including i) parametric trends in film, sheeting, and filament characteristics as a function of fluid rheology, delivery rates, and bell speed; ii) the impact of nonuniform film thicknesses on atomization performance; and iii) an understanding of spray composition via primary and secondary atomization. These findings result in coating design principles that are poised to improve energy- and cost-efficiency in a wide array of industrial and manufacturing settings.

Study on the calculational framework development of the advanced numerical reactor neutronics code SHARK

The SHARK program (Simulation-based High-fidelity Advanced Reactor physics Kit) is a high-fidelity heterogeneous neutronics code for the numerical reactor system being developed at the Nuclear Power Institute of China (NPIC). The program uses a Constructive Solid Geometry (CSG) framework to model various complex geometries. To enhance the flexibility and robustness during continuous development process, SHARK program attempts to support a rich set of methods, tools and library options within a unified general framework as a “toolkit”. For the multi-core clustered supercomputer architectures that are commonly used today, the SHARK program adopts a hybrid parallel strategy of MPI and OpenMP to achieve complementary advantages between them. In addition, the framework of SHARK program is designed with a true object-oriented manner. Through reasonable abstraction, inheritance and encapsulation, the maintainability and extensibility of the code are improved, and long-term team development is facilitated. Up to now, key modules for cross-section generation, heterogeneous transport calculation and microscopic depletion have been developed under the general frameworks. The main features of SHARK’s “resonance-transport-depletion” coupling system are elaborated in this paper, and some verification and validation (V&V) results in the current phase are presented and discussed.

Multiphysics Computing of Challenging Antenna Arrays Under a Supercomputer Framework

A parallel multiphysics simulation solver is developed to solve electromagnetic-thermal-mechanical coupling for some challenging large-scale antenna arrays. To achieve high scalability of supercomputer architectures, we reconstruct the preconditioned BiCGSTAB method and the non-overlapping domain decomposition method, so that the most resource-intensive matrix factorization steps can be performed in parallel independently within subdomains. The electromagnetic and thermal fields are solved separately, while coupled through the dissipated power and the temperature-dependent material parameters; after thermal steady state is reached, the mechanical simulation is stimulated subject to the temperature rise. The accuracy of electromagnetic-thermal coupling and thermal stress solution are first validated, and then the strong/weak parallel scalability experiments of the developed multiphysics solver are performed on supercomputer. Finally, an extremely challenging antenna array is simulated using the proposed solver, where to our best knowledge we bring the scale of multiphysics simulations excited by frequency-domain electromagnetic fields to the order of billion unknowns for the first time.

Characterization of Transmission Lines in Microelectronic Circuits Using the ARTEMIS Solver

Modeling and characterization of electromagnetic wave interactions with microelectronic devices to derive network parameters has been a widely used practice in the electronic industry. However, as these devices become increasingly miniaturized with finer-scale geometric features, computational tools must make use of manycore/GPU architectures to efficiently resolve length and time scales of interest. This has been the focus of our open-source solver, ARTEMIS (Adaptive mesh Refinement Time-domain ElectrodynaMIcs Solver), which is performant on modern GPU-based supercomputing architectures while being amenable to additional physics coupling. This work demonstrates its use for characterizing network parameters of transmission lines using established techniques. A rigorous verification and validation of the workflow is carried out, followed by its application for analyzing a transmission line on a CMOS chip designed for a photon-detector application. Simulations are performed for millions of timesteps on state-of-the-art GPU resources to resolve nanoscale features at gigahertz frequencies. The network parameters are used to obtain phase delay and characteristic impedance that serve as inputs to SPICE models. The code is demonstrated to exhibit ideal weak scaling efficiency up to 1024 GPUs and 84% efficiency for 2048 GPUs, which underscores its use for network analysis of larger, more complex circuit devices in the future.

Relating Reported Speedup to Attainable Speedup in HPC Applications Based on a Programmer's Expertise

High Performance Computing (HPC) refers to the processing of complex computations using enhanced computing power for quick results and better accuracy. The parallel computing features of supercomputers provide an application program to obtain significant speedup over the serial implementation of certain problems in scientific computing. An HPC application largely relies on human expertise i.e., knowledge of the underlying architecture, algorithm used to solve the problem, his expertise in parallelizing the sequential code efficiently. Hence the speedup is largely dependent on the programmers expertise. Therefore, an obtained speedup of a certain HPC application, may not always be the best possible speedup that is potentially possible on that supercomputer. So, there is always a need of finding out if the obtained speedup is the highest possible speedup for a given problem and on a given supercomputer architecture. Thats why its important to objectively rate a programmers ability to relate it to his capability to parallelize a given code. Although, online judge platforms have given ratings to programmers when they are tied to online judges (such as- Topcoder, Codeforce and Codechef), types of contests (long, short, national-level, international level and based on problem categories), and the programmers sustained interest in participating (frequency of participation), problems solved, problems attempted, maturity (in years and capability) and other factors. The ranking could also be misleading in some cases for example, in solving a difficult problem, a relatively new programmer may climb to a top position because of his familiarity with the category of the problem although he may have less submissions and experience. On the other hand, an experienced programmer can see a drop in his rank if he under-performs in one or more contests. Hence, there is no generalized way to rank a programmer based on a rating provided by an online judge platform. Therefore, we need a more sophisticated and reliable ranking model that can help us to evaluate the actual rating of a programmer. This work will involve researching existing ranking algorithms and develop a model that establishes a relationship between Programmers grade or level and the Maximum Attainable Speedup in a particular HPC application.

MemXCT: Design, Optimization, Scaling, and Reproducibility of X-Ray Tomography Imaging

This work extends our previous research entitled &#x201C;MemXCT: Memory-centric X-ray CT Reconstruction with Massive Parallelization&#x201D; that was originally published at SC19 conference (Hidayeto&#x011F;lu <i>et al.</i>, 2019) with reproducibility of the computational imaging performance. X-ray computed tomography (XCT) is regularly used at synchrotron light sources to study the internal morphology of materials at high resolution. However, experimental constraints, such as radiation sensitivity, can result in noisy or undersampled measurements. Further, depending on the resolution, sample size and data acquisition rates, the resulting noisy dataset can be in the order of terabytes. Advanced iterative reconstruction techniques can produce high-quality images from noisy measurements, but their computational requirements have made their use an exception rather than the rule. We propose a novel memory-centric approach that avoids redundant computations at the expense of additional memory complexity. We develop a memory-centric iterative reconstruction system, MemXCT, that uses an optimized SpMV implementation with two-level pseudo-Hilbert ordering and multi-stage input buffering. We evaluate MemXCT on various supercomputer architectures involving KNL and GPU. MemXCT can reconstruct a large (11K&#x00D7;11K) mouse brain tomogram in 10 seconds using 4096 KNL nodes (256K cores). The results presented in our original article at the SC19 were based on large-scale supercomputing resources. The MemXCT application was selected for the Student Cluster Competition (SCC) Reproducibility Challenge and evaluated on a variety of cloud computing resources by universities around the world in the SC20 conference. We summarize the results of the top-ranked SCC Reproducibility Challenge teams and identify the most pertinent measures for ensuring the reproducibility of our experiments in this article.

Runko: Modern multiphysics toolbox for plasma simulations

runko is a new open-source plasma simulation framework implemented in C++ and Python. It is designed to function as an easy-to-extend general toolbox for simulating astrophysical plasmas with different theoretical and numerical models. Computationally intensive low-level kernels are written in modern C++ taking advantage of polymorphic classes, multiple inheritance, and template metaprogramming. High-level functionality is operated with Python scripts. The hybrid program design ensures good code performance together with ease of use. The framework has a modular object-oriented design that allows the user to easily add new numerical algorithms to the system. The code can be run on various computing platforms ranging from laptops (shared-memory systems) to massively parallel supercomputer architectures (distributed-memory systems). The framework supports heterogeneous multiphysics simulations in which different physical solvers can be combined and run simultaneously. Here, we showcase the framework’s relativistic particle-in-cell (PIC) module by presenting (i) 1D simulations of relativistic Weibel instability, (ii) 2D simulations of relativistic kinetic turbulence in a suddenly stirred magnetically-dominated pair plasma, and (iii) 3D simulations of collisionless shocks in an unmagnetized medium.

GronOR: Scalable and Accelerated Nonorthogonal Configuration Interaction for Molecular Fragment Wave Functions.

GronOR is a program package for nonorthogonal configuration interaction calculations. Electronic wave functions are constructed in terms of antisymmetrized products of multiconfiguration molecular fragment wave functions. The computational complexity of the nonorthogonal methodologies implemented in GronOR applied to large molecular assemblies requires a design that takes full advantage of massively parallel supercomputer architectures and accelerator technologies. This work describes the implementation strategy and resulting performance characteristics. In addition to parallelization and acceleration, the software development strategy includes aspects of fault resiliency and heterogeneous computing. The program was designed for large-scale supercomputers but also runs effectively on small clusters and workstations for small molecular systems. GronOR is available as open source to the scientific community.

A massively parallel time-domain coupled electrodynamics–micromagnetics solver

We present a high-performance coupled electrodynamics–micromagnetics solver for full physical modeling of signals in microelectronic circuitry. The overall strategy couples a finite-difference time-domain approach for Maxwell’s equations to a magnetization model described by the Landau–Lifshitz–Gilbert equation. The algorithm is implemented in the Exascale Computing Project software framework, AMReX, which provides effective scalability on manycore and GPU-based supercomputing architectures. Furthermore, the code leverages ongoing developments of the Exascale Application Code, WarpX, which is primarily being developed for plasma wakefield accelerator modeling. Our temporal coupling scheme provides second-order accuracy in space and time by combining the integration steps for the magnetic field and magnetization into an iterative sub-step that includes a trapezoidal temporal discretization for the magnetization. The performance of the algorithm is demonstrated by the excellent scaling results on NERSC multicore and GPU systems, with a significant (59×) speedup on the GPU using a node-by-node comparison. We demonstrate the utility of our code by performing simulations of an electromagnetic waveguide and a magnetically tunable filter.

Towards hybrid supercomputing architectures

In light of recent work on combining control-flow and dataflow architectures on the same chip die, a new architecture based on an asymmetric multicore processor is proposed. The control-flow architectures are described as a most commonly used computer architecture today. Both multicore and manycore architectures are explained, as they are based on the same principles. A dataflow computing model assumes that data input flows through hardware as either a software or hardware dataflow implementation. In software dataflow, processors based on the control-flow paradigm process tasks based on their availability from the same queue (if there are any). In hardware dataflow architectures, the hardware is configured for a particular algorithm, and data input is streamed into the hardware, and the output is streamed back to the multicore processor for further processing. Hardware dataflow architectures are usually implemented with FPGAs. Hybrid architectures employ asymmetric multicore and manycore computer architectures that are based on the control-flow and hardware dataflow architecture, all combined on the same chip die. Advantages include faster processing time, lower power consumption (and heating), and less space needed for the hardware.