• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.776-787
• /
• 2000

Write requests which stored in disk cache of the RAID5 controller should be destaged to disk arrays according to the destage algorithm. As the response performance of the parallel IO request is being hit by the effect of the destage, several destage algorithms have been studied to enhance the performance of he RAID5 controller. Among the destage algorithms, the load balancing destage algorithm has better performance than other destage algorithms when system load is highly increased. But the load balancing destage algorithm gives priority to load balance among the disks in disk arrays, therefore, when some disks are affected by the very heavy system load caused by small data requests, the load balancing destage algorithm cannot enhance the performance of parallel IO requests effectively since it makes effort to maintain the load balance without the benefit of the locality of the write requests. This paper proposes a new RAID5 controller that applied reference-load balancing destage algorithm which decides the destage priority based on the reference history and load distribution of the disks. The simulation results show that RAID5 controller with the reference-load balancing destage algorithm has better performance than previous load balancing destage algorithm.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.11
• /
• pp.549-560
• /
• 2001

Cluster system provides attractive scalability in terms of compution power and memory size. With the advances in high speed computer network technology, cluster systems are becoming increasingly competitive compared to expensive MPPs (massively parallel processors). Load balancing is very important issue since an inappropriate scheduling of tasks cannot exploit the true potential of the system and can offset the gain from parallelization. In parallel processing program, it is difficult to predict the load of each task before running the program. Furthermore, tasks are interdependent each other in many ways. The dynamic load balancing algorithm, which evaluates each processor's load in runtime, partitions each task into the appropriate granularity and assigns them to processors in proportion to their performance in cluster systems. However, if the communication cost between processing nodes is expensive, it is not efficient for all nodes to attend load balancing process. In this paper, we restrict a processor that attend load balancing by the communication cost and the deviation of its load from the average. We simulate various models of the cluster system with parameters such as communication cost, node number, and range of workload value to compare existing load balancing methods with the proposed dynamic algorithms.

• Journal of Power Electronics
• /
• v.16 no.4
• /
• pp.1245-1255
• /
• 2016

The single resonant inverter is widely employed in typical inductive power transfer (IPT) systems to generate a high-frequency current in the primary side. However, the power capacity of a single resonant inverter is limited by the constraints of power electronic devices and the relevant cost. Consequently, IPT systems fail to meet high-power application requirements, such as those in rail applications. Total harmonic distortion (THD) may also violate the standard electromagnetic interference requirements with phase shift control under light load conditions. A power regulation approach with selective harmonic elimination is proposed on the basis of a parallel multi-inverter to upgrade the power levels of IPT systems and suppress THD under light load conditions by changing the output voltage pulse width and phase shift angle among parallel multi-inverters. The validity of the proposed control approach is verified by using a 1,412.3 W prototype system, which achieves a maximum transfer efficiency of 90.602%. Output power levels can be dramatically improved with the same semiconductor capacity, and distortion can be effectively suppressed under various load conditions.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.21 no.3
• /
• pp.200-206
• /
• 2016

In the microgrid, inverters for energy storage system are generally constructed in a parallel structure because of capacity expandability, convenience of system maintenance, and reliability improvement. Parallel inverters are required to provide stable voltage to the critical load in PCC and to accurately share the current between each inverter. Furthermore, when islanding occurs, the inverters should change its operating mode from grid-connected mode to stand-alone mode. However, during clearing time and control mode change, the conventional control method has a negative impact on the critical load, that is, severe fluctuating voltage. In this study, a parallel operation control method is proposed. This method provides seamless mode transfer for the entire transition period, including clearing time and control mode change, and has accurate current sharing between each inverter. The proposed control method is validated through simulation and experiment.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.3
• /
• pp.1048-1057
• /
• 2015

This paper proposes a sensorless speed control algorithm for parallel-connected dual Surface-mounted Permanent Magnet Synchronous Motors (SPMSMs) fed by a single inverter. For stable parallel operation of synchronous motors with a single inverter, each motor has to be constantly kept in the synchronization state regardless of load torque. If the master motor with the larger load is controlled, the synchronous state will be maintained. Therefore, detection of the master motor is essential. Conventionally, the master motor is determined by comparing the rotor position error from the relation between the back-EMF for torque angle and the flux position; consequently, the position sensor is deemed essential for finding the rotor position. The parallel sensorless speed control method proposed in this paper uses no position sensor, instead it compares the flux-axis current from the connection between the back-EMF for torque angle and current in unbalanced load conditions. The results of simulation and experiment conducted verify the efficacy of the proposed method.

• Proceedings of the KIEE Conference
• /
• 1998.07f
• /
• pp.2048-2050
• /
• 1998

An uninterruptible power supply(UPS) with parallel operation is used to increase the power capacity of the system or to secure higher reliability at critical loads. In the parallel operating system composed of the multiple UPSs, load-sharing, i.e. current balance control between them is key technique. Because of its low impedance and quick response characteristics, inverter output current changes very rapidly and thereby easily researches an overload condition. The difference between total load current divided by number of operating inverters and its own current is detected as unbalanced current. Then frequency and voltage are controlled to minimize the active component and the reactive component. A good performance of the proposed load-sharing technique is verified by experiments in the parallel operating system with two 40kVA UPSs.

• Journal of KIEE
• /
• v.10 no.1
• /
• pp.35-42
• /
• 2000

An equivalent load sharing control based on the frequency and voltage droop concept for parallel operation of two three-phase Uninterruptible Power Supply (UPS) systems with no control interconnection lines is presented in this paper. First of all, due to the use of active power and reactive power as control variables, the characteristics of output powers according to amplitude and phase differences between output voltages of two UPS systems are analyzed. Secondly, simulation results under different line impedance demonstrate the feasibility of the wireless parallel operation control. Finally, experiments are presented to verify the theoretical discussion with two three-phase 20kVA UPS systems employed TMS320C32, a kind of real time digital signal processor (DSP).

• Structural Engineering and Mechanics
• /
• v.45 no.6
• /
• pp.759-777
• /
• 2013

In this paper, a topology optimization method based on the element independent nodal density (EIND) is developed for continuum solids with multiple load cases and multiple constraints. The optimization problem is formulated ad minimizing the volume subject to displacement constraints. Nodal densities of the finite element mesh are used a the design variable. The nodal densities are interpolated into any point in the design domain by the Shepard interpolation scheme and the Heaviside function. Without using additional constraints (such ad the filtering technique), mesh-independent, checkerboard-free, distinct optimal topology can be obtained. Adopting the rational approximation for material properties (RAMP), the topology optimization procedure is implemented using a solid isotropic material with penalization (SIMP) method and a dual programming optimization algorithm. The computational efficiency is greatly improved by multithread parallel computing with OpenMP to run parallel programs for the shared-memory model of parallel computation. Finally, several examples are presented to demonstrate the effectiveness of the developed techniques.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.53 no.4
• /
• pp.255-261
• /
• 2004

This paper presents an improved droop method which minimizes the voltage droop of a parallel-connected power supply Conventionally, the droop method has been used to achieve a simple structure and no-interconnections among power sources. However, it has a trade-off between output voltage regulation and load sharing accuracy In this paper, the droop is minimized with a current and droop gain control using steady-stage estimation. The proposed method can achieve both good voltage regulation and good load sharing. A design example of two 10㎸, 100㎃ parallel modules is made and tested to verify the proposed current-sharing method.

• KIPS Transactions on Computer and Communication Systems
• /
• v.4 no.10
• /
• pp.337-342
• /
• 2015

We develop a new parallelization method for Sieve of Eratosthenes algorithm, which enhances both computation speed and energy efficiency. A pipeline scheduling is included for better load balancing after proper workload partitioning. They run on multicore CPUs with hybrid parallel programming model which uses both message passing and multithreading computation. Experimental results performed on both small scale clusters and a PC with a mobile processor show significant improvement in execution time and energy consumptions.