• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.497-522
• /
• 2018

Fully homomorphic encryption(FHE) scheme may be the best method to solve the privacy leakage problem in the untrusted servers because of its ciphertext calculability. However, the existing FHE schemes are still not being put into the practical applications due to their low efficiency. Therefore, it is imperative to find a more efficient FHE scheme or to optimize the existing FHE schemes so that they can be put into the practical applications. In this paper, we optimize GSW scheme by using the parallel computing, and finally we get a high-performance FHE scheme, namely PGSW scheme. Experimental results show that the time overhead of the homomorphic operations in new FHE scheme will be reduced manyfold with the increasing of processing units number. Therefore, our scheme can greatly reduce the running time of homomorphic operations and improve the performance of FHE scheme through sacrificing hardware resources. It can be seen that our FHE scheme can catalyze the development of FHE.

• Communications of the Korean Mathematical Society
• /
• v.10 no.3
• /
• pp.735-750
• /
• 1995

We present a domain decomposition algorithm for solving large sparse linear systems of equations arising from queuing networks. Such techniques are attractive since the problems in subdomains can be solved independently by parallel processors. Many of the methods proposed so far use some form of the preconditioned conjugate gradient method to deal with one large interface problem between subdomains. However, in this paper, we propose a "nested" domain decomposition method where the subsystems governing the interfaces are small enough so that they are easily solvable by direct methods on machines with many parallel processors. Convergence of the algorithms is also shown.lso shown.

• Journal of Korean Institute of Industrial Engineers
• /
• v.17 no.2
• /
• pp.131-142
• /
• 1991

The concept of criterion function is proposed as a framework for comparing the geometric and computational characteristics of various nonlinear programming approaches to linear programming such as the method of centers, Karmakar's algorithm and the gravitational method. Also, we discuss various computational issues involved in obtaining an efficient parallel implementation of these methods. Clearly, the most time consuming part in solving a linear programming problem is the direction finding procedure, where we obtain an improving direction. In most cases, finding an improving direction is equivalent to solving a simple optimization problem defined at the current feasible solution. Again, this simple optimization problem can be seen as a least squares problem, and the computational effort in solving the least squares problem is, in fact, same as the effort as in solving a system of linear equations. Hence, getting a solution to a system of linear equations fast is very important in solving a linear programming problem efficiently. For solving system of linear equations on parallel computing machines, an iterative method seems more adequate than direct methods. Therefore, we propose one possible strategy for getting an efficient parallel implementation of an iterative method for solving a system of equations and present the summary of computational experiment performed on transputer based parallel computing board installed on IBM PC.

• 제어로봇시스템학회:학술대회논문집
• /
• 1988.10a
• /
• pp.1-5
• /
• 1988

The realization of high quality robot control needs the improvement of computing speed of controller. In this paper, parallel processing method is considered for this purpose. A S/W algorithm for task scheduling is developed first, and then, an appropriate H/W structure is proposed. This scheme is applied to calculate inverse kinematics of PUMA robot. The simulation results show that the computing time when using three 8086/87's is reduced to 4.23 msec compared to 10 msec in case using one 8086/87.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.2 no.1
• /
• pp.106-115
• /
• 1994

A parallel integration approach is proposed for real-time simulation of controlled mechanical systems. The proposed approach, which employs the dual-rate integration method in a parallel computing environment, is developed to deal with stiffness and high frequency characteristics of the controlled mechanical systems effectively. Numerical experiments are performed to demonstrate the effectiveness of the approach in shared memory multiprocessors, Alliant FX/8 and Alliant FX/80.

• Journal of applied mathematics & informatics
• /
• v.6 no.1
• /
• pp.279-289
• /
• 1999

Given n points in the plane the planar convex hull prob-lem in that of finding which of these points belong to the perimeter of the smallest convex region (a polygon) containing all n points. Here we suggest two kinds of methods. First we present a new sequential method for constructing the pla-nar convex hull O(1.5n) time in the quadratic decision tree model. Second using the sequential method we suggest a new parallel algo-rithm which solve the planar convex hull O(1.5n/p) time on a maspar Machine (CREW-PRAM) with O(n) processors. Also when we run on a maspar Machine we achieved a 37. 156-fold speedup with 64 pro-cessor.

• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.1040-1046
• /
• 2009

Thanks to the recent development of computing technology, the various forms of high-performance computers are available. Among them, the parallel-clustering CPU machines are realized for the high performance computing. These supercomputers (cluster computers) can be applied to various industries due to the advantages of lower price. Especially in the field of numerical flow simulation, use of supercomputers can produce results quickly, and various engineering problems can be reviewed effectively case by case. In this paper, an application of supercomputers (cluster computers) were examined for railroad industry of fire flow simulation by using parallel computational method. It make sure that the supercomputers are very useful tools for railroad engineering.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.12
• /
• pp.2109-2115
• /
• 2007

The wind generators have been installed with high output power to increase the energy production and efficiency. Hence, Optimal design of the direct-driven PM wind generator, coupled with F.E.M(Finite Element Method) and Genetic Algorithm(GA), has been performed to maximize the Annual Energy Production(AEP) over the whole wind speed characterized by the statistical model of wind speed distribution. Particularly, the parallel computing via internet web service has been applied to loose excessive computing times for optimization. The results of the optimal design of Surface-Mounted Permanent Magnet Synchronous Generator(SPMSG) are compared with each other candidates to verify the usefulness of the maximizing AEP model.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.14 no.3
• /
• pp.287-298
• /
• 2001

Analysis of nonlinear behavior of prestressed concrete frame structures on PC is a time-consuming computing job if the problem size increase to a certain degree. Cluster system has emerged as one of promising computing environments due to its good extendibility, portability, and cost-effectiveness, comparing it with high-end work-stations or servers. In this paper, a parallel nonlinear analysis procedure of prestressed concrete frame structure is presented using cluster computing. Cluster system is configured with readily available pentium III class PCs under Win98 or Linux and fast ethernet. Parallel computing algorithms on element-wise processing parts including the calculation of stiffness matrix, element stresses and determination of material states, check of material failure and calculation of unbalanced loads are developed using MPL. Validity of the method is discussed through typical numerical examples. For the case of 4 node system, maximum speedup is 3.15 and 3.74 for Win98 and Linux, respectively. Important issues for the efficient use of cluster computing system based un PCs and ethernet are addressed.

• Structural Engineering and Mechanics
• /
• v.38 no.4
• /
• pp.405-416
• /
• 2011

An efficient two-level domain decomposition parallel algorithm is suggested to solve large-DOF structural problems with nonlinear material models generating unsymmetric tangent matrices, such as a group of plastic-damage material models. The parallel version of the stabilized bi-conjugate gradient method is developed to solve unsymmetric coarse problems iteratively. In the present approach the coarse DOF system is solved parallelly on each processor rather than the whole system equation to minimize the data communication between processors, which is appropriate to maintain the computing performance on a non-supercomputer level cluster system. The performance test results show that the suggested algorithm provides scalability on computing performance and an efficient approach to solve large-DOF nonlinear structural problems on a cluster system.