Heterogeneous Cluster Architectures Research Articles

Energy saving of cluster computing by CPU frequency Tuning using genetic algorithm

Dynamic voltage and frequency scaling (DVFS) is a tool used primarily to decrease computer processor energy consumption by lowering its operational frequency. Their only downside is that they distract from the efficiency of parallel applications while operating on parallel platforms. In a heterogeneous cluster architecture, however, a genetic algorithm is being implemented and applied to model the best trade-off between energy-saving and parallel application performance degradation. The proposed algorithm selects the best frequency vector in order to accomplish these objectives by providing the same compromise. So, the objective function of the genetic algorithm at the same time gives limited energy consumption and minimum decreases in performance. The SimGrid simulator will be used for all experiments. The suggested algorithm saves the average energy by (20 %) and the application performance degrades to the limit (0.15 %).

Selecting Best CPU frequency for energy saving in cluster using genetic algorithm

Dynamic voltage and frequency scaling (DVFS) is a technique mainly used for reducing the consumed energy of computer’s processor. Its only drawback detracting the performance of parallel application when executing them over parallel platform. However, genetic algorithm is introduced and applied in a heterogeneous cluster architecture for modeling the best trade-off between the energy saving and performance degradation of the parallel application in the same time. The suggested algorithm selects the best frequencies vector to achieve that targets by offering equal trade-off between them. The genetic algorithm simultaneously gives minimum energy consumption and minimal performance degradation via its objective function. All experiment will apply using SimGrid simulator. The experiments show that the algorithm reduces the energy consumption of the message passing application by (24% with 8000 problem size and 22% for 4000 problem size) and in almost cases improve the application performance up to 3%.

Massively parallel computation of accurate densities for N-body dark matter simulations using the phase-space-element method

This paper presents an accurate density computation approach for large dark matter simulations, based on a recently introduced phase-space tessellation technique and designed for massively parallel, heterogeneous cluster architectures. We discuss a memory efficient construction of an oct-tree structure to sample the mass densities with locally adaptive resolution, according to the features of the underlying tetrahedral tessellation. We propose an efficient GPU implementation for the computationally intensive operation of intersecting the tetrahedra with the cubical cells of the deposit grid, that achieves a speedup of almost an order of magnitude compared to an optimized CPU version. We discuss two dynamic load balancing schemes - the first exchanges particle data between cluster nodes and deposits all tetrahedra for each block of the grid structure on single nodes, whereas the second approach uses global reduction operations to obtain the total masses. We demonstrate the scalability of our algorithms for up to 256 GPUs and TB-sized simulation snapshots, resulting in tessellations with over 400 billion tetrahedra.

Data Partitioning Strategy of GPU Heterogeneous Clusters Based on Learning

With the rapid progress of computational science and computer simulation ability, a lot of properties can be predicted by the powerful ability of parallel computation before the actual research and development. With the development of high performance computer architecture, GPU is more and more widely used in high performance computation field as an emerging architecture, and a growing number of computations use GPU heterogeneous cluster architecture. However, how to partition workload and map to computing resource has always been the focus and difficult point. In the current study of GPU, according to the problems of the computing power provided by each node and the cluster hardware architecture which the application programmers don't understand, some partitioning strategies will result in serious load imbalance problem. Aimed at the complexity brought by the different computing ability of the nodes of GPU clusters, this paper proposes a GPU data partitioning strategy of heterogeneous clusters based on learning. It collects the states of each node in the process of running a program, and then estimates the calculation ability of each node dynamically, so as to guide the data partitioning. Actual testing results show that, this strategy allocates different tasks to nodes based on computing ability to ensure load balancing among nodes, so as to improve the execution performance of CUDA programs on heterogeneous GPU clusters and it laid a solid foundation for efficient computing on heterogeneous GPU clusters.

Параллельная реализация алгоритма разреженного QR разложения для прямоугольных верхних квазитреугольных матриц со структурой разреженности типа вложенных сечений

The paper considers parallel MPI+threads+SIMD implementation of the algorithm for computing sparse QR decomposition of a specially ordered rectangular matrix. Decomposition is based on block sparse Householder transformations. The algorithm starts with independent parallel QR decompositions for sets of matrix rows; and then, according to the computations tree, the QR decomposition is performed for matrices, combined with elements of R factors of rows decompositions. The results of numerical experiments for test problems show efficiency of the parallel implementation. The algorithm can also be efficiently implemented on heterogeneous cluster architectures with GPGPU accelerators.

Heterogeneous cluster architectures and applications

Heterogeneous cluster architectures and applications

Computational gas dynamics in a wide range of Mach number on heterogeneous cluster architecture

The specific features of numerical method for gas dynamics problems are considered in this paper. The viscous flows are considered in a wide range of Mach number. The numerical method is based on conservation laws applied on unstructured heterogeneous mesh with further integration on complex cluster architectures. Application of unstructured mesh allows one to perform calculation of geometrically complex 3D objects with strict conservation properties. The developed numerical method includes multiple flux solvers (Godunov's, HLLC, Roe-Fix, AUSM, etc), first and second order spatial approximation, and explicit, semi-implicit or implicit first order time schemes. The method utilizes heterogeneous GPU HPC clusters. A wide range of problems is used for verification of the developed computational complex. All results on these problems had a good agreement with reference data.

Efficient grouping-based mapping and scheduling on heterogeneous cluster architectures

Heterogeneous clusters of computers usually provide high computing power for large-scale applications at the expense of large cost. And there are two challenges currently faced by researchers. One is how to map large applications, modeled by Directed Acyclic Graphs (DAG), to heterogeneous architectures with minimal cost. The other is how to schedule tasks on each cluster to further decrease the total cost.This paper proposes an Integer Linear Programming (ILP) formulation to achieve optimal results. In order to improve the efficiency for ILP, the problem size is first reduced by Safe Graph Grouping (SGG). SGG guarantees the reduced graph to be a DAG to avoid dead lock during the scheduling. Then Time Constrained Local Scheduling (TCLS) is used to further reduce total cost for the input task graph. Experimental results show that the cooperation of SGG and ILP can reduce the total cost by 11.46% compared with existing mapping techniques.

Parallel text searching applications on a heterogeneous cluster architecture

In this paper, we propose four text searching implementations on a cluster of heterogeneous workstations using MPI message passing library. The first three parallel implementations are based on the static and dynamic master-worker methods. Further, we propose a hybrid parallel implementation that combines the advantages of static and dynamic parallel methods in order to reduce the load imbalance and communication overhead. We test these parallel implementations and present experimental results for different text sizes and number of workstations. We also propose a performance prediction model of the four implementations that agrees well with our experimental measurements.

List scheduling: extension for contention awareness and evaluation of node priorities for heterogeneous cluster architectures

In the area of static scheduling, list scheduling is one of the most common heuristics for the temporal and spatial assignment of a directed acyclic graph (DAG) to a target system. As most scheduling heuristics, list scheduling assumes fully connected homogeneous processors and ignores contention on the communication links. This article extends the list scheduling heuristic for contention aware scheduling on heterogeneous arbitrary architectures. The extension is based on the idea of scheduling edges to links, likewise the scheduling of nodes to processors. Based on this extension, we compare eight priority schemes for the node order determination in the first phase of list scheduling. Random graphs are generated and scheduled with the different schemes to homogenous and heterogeneous parallel systems from the area of cluster computing. Apart from identifying the best priority scheme, the results give new insights in contention aware DAG scheduling. Moreover, we demonstrate the appropriateness of our extended list scheduling for homogeneous and heterogeneous cluster architectures.