• The Transactions of the Korea Information Processing Society
• /
• v.7 no.4
• /
• pp.1201-1210
• /
• 2000

In order to tolerate faults which may occur during the execution of distributed tasks in high-performance parallel computer systems, tasks are duplicated on different processors. In this paper, by utilizing the task duplication based scheduling algorithm, a new task scheduling algorithm which duplicates each task on more than two different processors with the minimum schedule length is presented, and the number of processors required for the duplication is analyzed with the ratio of communication cost to computation time and the workload of the system. A simulation with various task graphs reveals that the number of processors required for the full-duplex fault-tolerant task scheduling with the obtainable minimum schedule length increases about 30% to 75% when compared with that of the task duplication based scheduling algorithm.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.289-293
• /
• 2000

We propose two new algorithms for parallelism-independent scheduling. The machine code generated from the compiler using these algorithms in its scheduling phase is parallelism-independent code, executable in minimum time regardless of the number of the processors in the parallel computer. Our new algorithms have the following phases: finding the minimum number of processors on which the program can be executed in minimal time, scheduling by an heuristic algorithm for this predefined number of processors, and serialization of the parallel schedule according to the earliest start time of the tasks. At run time tasks are taken from the serialized schedule and assigned to the processor which allows the earliest start time of the task. The order of the tasks decided at compile time is not changed at run time regardless of the number of the available processors which means there is no out-of-order issue and execution. The scheduling is done predominantly at compile time and dynamic scheduling is minimized and diminished to allocation of the tasks to the processors. We evaluate the proposed algorithms by comparing them in terms of schedule length to the CP/MISF algorithm. For performance evaluation we use both randomly generated DAGs (directed acyclic graphs) and DACs representing real applications. From practical point of view, the algorithms we propose can be successfully used for scheduling programs for in-order superscalar processors and shared memory multiprocessor systems. Superscalar processors with any number of functional units can execute the parallelism-independent code in minimum time without necessity for dynamic scheduling and out-of-order issue hardware. This means that the use of our algorithms will lead to reducing the complexity of the hardware of the processors and the run-time overhead related to the dynamic scheduling.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.2
• /
• pp.279-287
• /
• 1993

In this paper, double laxity threshold MLT(Minimum Laxity Threshold) algorithm and double queue threshold QLT(Queue Length Threshold) algorithm are proposed as DPS(Dynamic Priority Scheduling) techniques for advanced processing of multiple class traffics. Also, the performance of the proposed algorithms is analyzed by a computer simulation. According to the simulation results, it can be shown that the proposed double laxity threshold ML T algorithm advances the processing performance versus ML T algorithm for 2 or more classes delay sensitive traffics, and that double queue length threshold QL T algorithm provides more efficient performance than QL T for 2 or more classes of non real time traffics.

• The KIPS Transactions:PartA
• /
• v.11A no.7 s.91
• /
• pp.511-520
• /
• 2004

A peer-to-peer model is very useful in solving the server link bottleneck problem of a client-server model. In this work, we discuss the problems of distributing multimedia content over peer-to-peer network. We focus on two problems in peer-to-peer media content distribution systems. The first is the transmission scheduling of the media data for a multi-source streaming session. We present a sophisticated scheduling scheme called fixed-length slotted scheduling, which results in minimum buffering delay. The second problem is on the fast distribution of media content in the peer-to-peer system that is self-growing. We propose a mechanism accelerating the speed at which the system's streaming ca-pacity increases, called FAST.

• Journal of Communications and Networks
• /
• v.11 no.5
• /
• pp.489-499
• /
• 2009

We describe the fundamental issue of scheduling the allocation of wireless network resources and provide several formulations of the associated problems. The emphasis is on scheduling transmission attempts. We place this problem in the context of existing approaches, like information theoretic and traditional network theoretic ones, as well as novel avenues that open up the possibility of addressing this issue for non-stationary and non-ergodic environments. We summarize concrete recent results for specific special cases that include unicast and multicast traffic, different objective functions, and reduced complexity versions of the problem. We conclude with some thoughts for future work. We identify and single out the cross-layer nature of the problem and include a simple physical-layer criterion in what is mostly a medium access control (MAC) problem.

• Journal of Communications and Networks
• /
• v.6 no.2
• /
• pp.147-155
• /
• 2004

In this paper, we propose an efficient sequencing technique, namely minimum Row time scheduling (MFTS), to manage variable-Iength message transmissions for single-hop passive starcoupled WDM optical networks. By considering not only the message length but also the state of the receivers and the tuning latency, the proposed protocol can reduce the average delay of the network greatly. This paper also introduces a new channel assignment technique latency minimizing scheduling (LMS), which aims to reduce the scheduling latency. We evaluate the proposed algorithm, using extensive discrete-event simulations, by comparing its performance with shortest job first (SJF) algorithm. We find that significant improvement in average delay could be achieved by MFTS algorithm. By combining the proposed message sequencing technique with the channel selection technique, the performance of the optical network could be further improved.

• Journal of KIISE:Information Networking
• /
• v.28 no.3
• /
• pp.447-453
• /
• 2001

In this paper, we consider the scheduling of SE(switching element)-disjoint multicasting in photonic Banyan-type switching networks constructed with directional couplers. This ensures that at most, one connection holds each SE in a given time thus, neither crosstalk nor blocking will arise in the network. Such multicasting usually takes several routing rounds hence, it is desirable to keep the number of rounds(i.e., scheduling length) to a minimum. We first present the necessary and sufficient condition for connections to pass through a common SE(i.e., make crosstalk) in the photonic Banyan-type networks capable of supporting one-to-many connections. With definition of uniquely splitting a multicast connection into distinct subconnections, the crosstalk relationship of a set of connections is represented by a graph model. In order to analyze the worst case crosstalk we characterize the upper bound on the degree of the graph. The successor paper(Part II)[14] is devoted to the scheduling algorithm and the upper bound on the scheduling length. Comparison with related results is made in detail.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.7
• /
• pp.1785-1796
• /
• 1998

This paper is concerned with the relationships between optimal number of wavelengths and tuning time in time division multiplexing(TDM) scheduling on wavelength division multiplexing(WDM) optical passive star networks. We assume that the traffic is nonuniform and each node has a tunable transmitter and a fixed receiver. Each node transmits spckets to all other nodes. Most of the earlier protocols on TDM based scheduling for WDM star networks [7, 8, etc.] use whold given wavelength chnnels. But in this paper, we investigate the optimal number of wavelengths that yidels minimum frame length when tuning time exists. It appears within the availble number of wavelengths that yields minimum frame length when tuning time exists. It appears within the available number of wavelengths. We analyze the relationships between optimal number of wavelengths and tuning time by experiments. We analyze the reationships between optimal number of wavelengths and tuning time by experiments. We also discuss on the possibility of reduction of frame length by increasing the number of nodes trans-mitters and receivers.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.25 no.2
• /
• pp.92-101
• /
• 2002

The Job Shop Scheduling Problem(JSSP) is concerned with schedule of m different machines and n jobs where each job consists of a chain of operations, each of which needs to be processed during an uninterrupted time period of a given length on a given machine. The purpose of this paper is to develop the efficient heuristic method for solving the minimum makespan problem of the large scale job shop scheduling. The proposed heuristic method is based on a Tabu Search(TS) and on a Constraint Satisfaction Technique(CST). In this paper, ILOG libraries is used to embody the job shop model, and a CST is developed for this model to generate the increased solution. Then, TS is employed to overcome the increased search time of CST on the increased problem size md to refine the next-current solution. Also, this paper presents the new way of finding neighbourhood solution using TS. On applying TS, a new way of finding neighbourhood solution is presented. Computational experiments on well known sets of MT and LA problem instances show that, in several cases, our approach yields better results than the other heuristic procedures discussed In literature.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.6
• /
• pp.47-53
• /
• 2016

This paper deals with the capacity constrained time slot assignment problem(CTSAP) that a satellite switches to traffic between $m{\times}n$ ground stations using on-board $k{\leq}_{min}\{m,n\}$ k-transponders switching modes in SS/TDMA time-division technology. There was no polynomial time algorithm to solve the optimal solution thus this problem classified by NP-hard. This paper suggests a heuristic algorithm with O(mn) time complexity to solve the optimal solution for this problem. Firstly, the proposed algorithm selects maximum packet lengths of $\({mn \atop c}\)$ combination and transmits the cut of minimum packet length in each switching mode(MSMC). In the case of last switching mode with inefficient transmission, we applies a compensation strategy to obtain the minimum number of switching modes and the minimum makespan. The proposed algorithm finds optimal solution in polynomial time for all of the experimental data.