• The Transactions of the Korea Information Processing Society
• /
• v.2 no.3
• /
• pp.288-298
• /
• 1995

Load balancing attempts to improve the performance of a distributed computing system by transferring some of the workload of a congested site to others. A load balancing scheme that supports both the source-initiated and the server- initiated load balancing is proposed in this paper. It can model both the m/m/1 queue(no load balancing) and the m/m/n queue(perfect load balancing) as the extreme State variables are replicated into every site, and copy consistency constraints are relaxed more weakly. We propose weak copy consistency constraints which correlate the outdated state information to that of the current state. We also derive sufficient conditions under which each scheduler can guarantee the load balancing without communication Using this model, the problem of load balancing is converted to that of maintaining the consistency of states and communication overhead becomes less than that of the bidding algorithm. The scheme is simulated by event tracing, compared to the two extreme cases and the bidding algorithm. The results show that the mean response time and the number of messages are reduced by 0-35% and 40-100% respectively, compared with the bidding algorithm. Finally the effects of some system parameters are described.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.1
• /
• pp.30-37
• /
• 2016

A multi-server queueing system with an infinite buffer and impatient customers is analyzed. The system operates in the finite state Markovian random environment. The number of available servers, the parameters of the batch Markovian arrival process, the rate of customers' service, and the impatience intensity depend on the current state of the random environment and immediately change their values at the moments of jumps of the random environment. Dynamics of the system is described by the multi-dimensional asymptotically quasi-Toeplitz Markov chain. The ergodicity condition is derived. The main performance measures of the system are calculated. Numerical results are presented.

• Journal of Korean Society of Transportation
• /
• v.19 no.3
• /
• pp.101-113
• /
• 2001

Since late 1970s. one of the principal research areas in transportation problem is dynamic traffic assignment (DTA). Although many models have been developed regarding DTA, yet they have some limits of describing real traffic patterns. This reason comes from the fact that DTA model has the time varying constraints such as state equation, flow propagation constraint, first in first out(FIFO) rule and queuing evolution. Thus, DTA model should be designed to satisfy these constraints as well as dynamic route choice condition, dynamic user equilibrium. In this respect, link-based DTA models have difficulty in satisfying such constraints because they have to satisfy the constraints for each link, while path-based DTA models may easily satisfy them. In this paper we develop a path-based DTA model. The model includes point queue theory to describe the queue evolution and simulation loading method for depicting traffic patterns in more detail. From a numerical test, the model shows promising results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10c
• /
• pp.201-203
• /
• 2000

VOQ 방법은 입력 큐잉의 저 비용과 출력 큐잉의 고성능이라는 장점을 결합한 효과적인 큐잉방법이다. VOQ를 효과적으로 사용하기 위해서는 셀의 순서를 스케쥴 해주는 스케쥴링 알고리즘이 필요하다. 본 논문에서는 기존의 iSlip, RRM의 round-robin based priority 방식을 따르면서 큐의 현재 상태에 따라서 스케쥴링을 달리하는 알고리즘을 제시하였다. 이 알고리즘에서는 큐에 대한 threshold값을 설정하고 큐에 저장하고 있는 셀의 수가 threshold값 보다 클 경우 그 큐의 셀을 연속해서 포워딩 하도록 하였다. 시뮬레이션을 통해 본 논문에서 제시한 알고리즘과 기존의 iSlip 알고리즘과의 성능을 패킷의 지연시간 측면에서 비교해 보았다.

• Journal of applied mathematics & informatics
• /
• v.2 no.2
• /
• pp.21-46
• /
• 1995

Congestion and memory occupancy in computer system may be reduced further if new jobs are admitted only when the num-ber of jobs queued at CPU is below CPU run queue cutoff (RQ). In this paper we prove that response time of a job is invariant with respect to RQ if jobs do not communicate each other. We also demonstrate this invariance property numerically using marix-geometric methods and present an approximate method for the delay due to context switch-ing under time slicing. The approximation suggests that time slicing with constant overhead yields a throughput similar to an FCFS system without overhead.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.22 no.2
• /
• pp.45-65
• /
• 1997

We build a queueing model for buffered leaky bucket system. First, we set up system equations and them calculate the steady-state probabilities at an arbitrary time epoch by recursive method. We derive the mean waiting time and the mean number of cells in the input buffer, and evaluate the performance of the buffered leaky bucket system to find the optimal queue capacity and token generation rate that meet the quality of service(QoS).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.5
• /
• pp.1384-1396
• /
• 1998

This paper ivestigates the performance of a relative rate (RR) switch algorithm for the rate-based available bit rate (ABR) flow control in asynchronous transfer mode (ATM) networks. A RR switch may notify the network congestion status to the source by suing the congestion indication (CI) bit or no increase (NI)bit in the backward RM (BRM) cells. A RR switch can be differently implemented according to the congestion detectio and notification methods. In this paper, we propose three implementation schemes for the RR switch with different congestion detection and notification methods, and analyze the allowed cell rate (ACR) of a source and the queue length of a switch in steady state. In addition, we derive the upper and lower bounds for the maximum and minimum queue lengths for each scheme respectively, and evaluate the effects of the ABR parameter values on the queue length. Furthermore, we suggest the range of the rage increase factor (RIF) and rate decrease factor (RDF) parameter values which can prevent buffer overflow and underflow at a switch.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.1
• /
• pp.89-99
• /
• 2020

In this paper, we consider a dyadic server control policy that combines workload control and single vacation. Customer arrives at the system with Bernoulli process, waits until his or her turn, and then receives service on FCFS(First come first served) discipline. If there is no customer to serve in the system, the idle single server spends a vacation of discrete random variable V. If the total service times of the waiting customers at the end of vacation exceeds predetermined workload threshold D, the server starts service immediately, and if the total workload of the system at the end of the vacation is less than or equal to D, the server stands by until the workload exceeds threshold and becomes busy. For the discrete-time Geo/G/1 queueing system operated under this dyadic server control policy, an idle period is analyzed and the steady-state queue length distribution is derived in a form of generating function.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.8
• /
• pp.619-630
• /
• 2003

We present a method of runtime configuration scheduling in reconfigurable SoC design. As a model of computation, we use a popular formal model of computation, hierarchical FSM (HFSM) with synchronous dataflow (SDF) model, in short, HFSM-SDF model. In reconfigurable SoC design with HFSM-SDF model, the problem of configuration scheduling becomes challenging due to the dynamic behavior of the system such as concurrent execution of state transitions (by AND relation), complex control flow (HFSM), and complex schedules of SDF actor firing. This makes it hard to hide configuration latency efficiently with compile-time static configuration scheduling. To resolve the problem, it is necessary to know the exact order of required configurations during runtime and to perform runtime configuration scheduling. To obtain the exact order of configurations, we exploit the inherent property of HFSM-SDF that the execution order of SDF actors can be determined before executing the state transition of top FSM. After obtaining the order information and storing it in the ready configuration queue (ready CQ), we execute the state transition. During the execution, whenever there is FPGA resource available, a new configuration is selected from the ready CQ and fetched by the runtime configuration scheduler. We applied the method to an MPEG4 decoder and IS95 design and obtained up to 21.8% improvement in system runtime with a negligible overhead of memory usage.

• Science of Emotion and Sensibility
• /
• v.7 no.4
• /
• pp.67-75
• /
• 2004

In this paper, we analyze previous flow control algorithm for serving ABR traffic, and then propose the algorithm which calculate fair transmission rate and control efficiently ABR traffic using VS/VD switch which has an effect on PNNI environment with long delay. For getting the transmission rate, the proposed algorithm use FMMRA as ER scheme which calculate exactly the fair share. And, in large delay state, we can obtain fair cell sharing by calculating transmission rates which obtained by transmitted queue length information of VD to VS for serve and drain cell in queue Through the computer simulation, we evaluate the performance of proposed algorithm, According to the results, the proposed algorithm shows good performance.