相关论文: Dynamic Load Balancing for Compressible Multiphase…
Parallel iterative applications often suffer from load imbalance, one of the most critical performance degradation factors. Hence, load balancing techniques are used to distribute the workload evenly to maximize performance. A key challenge…
Parallel applications with irregular and time-varying workloads often suffer from load imbalance. Dynamic load balancing techniques address this challenge by redistributing work during execution. We present a new type of distributed…
Traffic load-balancing in datacenters alleviates hot spots and improves network utilization. In this paper, a stable in-network load-balancing algorithm is developed in the setting of software-defined networking. A control plane configures…
Load balancing is the process of improving the Performance of a parallel and distributed system through is distribution of load among the processors [1-2]. Most of the previous work in load balancing and distributed decision making in…
In this paper, we derive and investigate approaches to dynamically load balance a distributed task parallel application software. The load balancing strategy is based on task migration. Busy processes export parts of their ready task queue…
In-Memory Computing (IMC) represents a paradigm shift in deep learning acceleration by mitigating data movement bottlenecks and leveraging the inherent parallelism of memory-based computations. The efficient deployment of Convolutional…
This paper presents a novel methodology for the direct numerical modeling and simulation of turbulent flows. The kinetic model equation is firstly extended to turbulent flow with the account of coupled evolution of kinetic, thermal, and…
The Combustion Toolbox (CT) is a newly developed open-source thermochemical code designed to solve problems involving chemical equilibrium for both gas- and condensed-phase species. The kernel of the code is based on the theoretical…
Chemistry Foundation Models (CFMs) that leverage Graph Neural Networks (GNNs) operating on 3D molecular graph structures are becoming indispensable tools for computational chemists and materials scientists. These models facilitate the…
Load balancing algorithms play critical roles in systems where the workload has to be distributed across multiple resources, such as cores in multiprocessor system, computers in distributed computing, and network links. In this paper, we…
Using tiny, equal-sized tasks (Homogeneous microTasking, HomT) has long been regarded an effective way of load balancing in parallel computing systems. When combined with nodes pulling in work upon becoming idle, HomT has the desirable…
We demonstrate neural-network runtime prediction for complex, many-parameter, massively parallel, heterogeneous-physics simulations running on cloud-based MPI clusters. Because individual simulations are so expensive, it is crucial to train…
Turbulent transport remains one of the principal obstacles to achieving efficient magnetic confinement in fusion devices. Two of the dominant drivers of the turbulence are microscale instabilities fuelled by electron- and ion-temperature…
The accuracy obtained with CFD and process simulations of flotation critically depends on the quality and robustness of the underlying models for the non-resolved sub-processes. An important issue in flotation is the collision between…
While load balancing in distributed-memory computing has been well-studied, we present an innovative approach to this problem: a unified, reduced-order model that combines three key components to describe "work" in a distributed system:…
In this work, we present first-of-their-kind nonlinear local gyrokinetic simulations of electromagnetic turbulence at mid-radius in the burning plasma phase of the conceptual high-$\beta$, reactor-scale, tight-aspect-ratio tokamak STEP…
Analysis of compressible turbulent flows is essential for applications related to propulsion, energy generation, and the environment. Here, we present BLASTNet 2.0, a 2.2 TB network-of-datasets containing 744 full-domain samples from 34…
To faithfully simulate ITER and other modern fusion devices, one must resolve electron and ion fluctuation scales in a five-dimensional phase space and time. Simultaneously, one must account for the interaction of this turbulence with the…
Modern supercomputers allow the simulation of complex phenomena with increased accuracy. Eventually, this requires finer geometric discretizations with larger numbers of mesh elements. In this context, and extrapolating to the Exascale…
Streaming applications frequently encounter skewed workloads and execute on heterogeneous clusters. Optimal resource utilization in such adverse conditions becomes a challenge, as it requires inferring the resource capacities and input…