• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.246-254
• /
• 1997

This paper proposes an improved scheduling model and algorithm which threads in the fixed-priority policy have faster response time and turnaround time than existing ones in SRT system. Sub run queue is added into the system and is used for the quick execution and aging of threads in the fixed-priority policy. Using simulation two method are compared. The results shows that the new scheduling model gets threads in the fixed-priority policy to run faster than existing ones.

• Korean Management Science Review
• /
• v.31 no.3
• /
• pp.13-26
• /
• 2014

This paper considers an integrated one-vendor multi-buyer production-inventory model where the vendor manufactures multiple products in lot at their associated finite production rates. In the model, it is allowed for each product to be shipped in lot to the buyers even before the whole product production is not completed yet. Each product lot is dispatched to the associated buyer in a number of shipments. The buyers consume their products at fixed rates. The objective is to the production and shipment schedules in the integrated system, which minimizes the total cost per unit time. The total cost consists of production setup cost, inventory holding cost and shipment cost. For the model, an iterative optimal solution procedure with shipment consolidation policy incorporated. It is then tested through numerical experiments to show how efficient and effective the shipment consolidation policy is.

• Journal of Computing Science and Engineering
• /
• v.8 no.1
• /
• pp.43-56
• /
• 2014

Semi-partitioned scheduling is a new approach for allocating tasks on multiprocessor platforms. By splitting some tasks between processors, semi-partitioned scheduling is used to improve processor utilization. In this paper, a new semi-partitioned scheduling algorithm called SS-DRM is proposed for multiprocessor platforms. The scheduling policy used in SS-DRM is based on the delayed rate monotonic algorithm, which is a modified version of the rate monotonic algorithm that can achieve higher processor utilization. This algorithm can safely schedule any system composed of two tasks with total utilization less than or equal to that on a single processor. First, it is formally proven that any task which is feasible under the rate monotonic algorithm will be feasible under the delayed rate monotonic algorithm as well. Then, the existing allocation method is extended to the delayed rate monotonic algorithm. After that, two improvements are proposed to achieve more processor utilization with the SS-DRM algorithm than with the rate monotonic algorithm. According to the simulation results, SS-DRM improves the scheduling performance compared with previous work in terms of processor utilization, the number of required processors, and the number of created subtasks.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.04a
• /
• pp.208-215
• /
• 1995

Numerous solution methods for scheduling problems such as part dispatching problem, operation sequence problem have been suggested as a means to be embedded in hierarchical or centralized shop floor control. Under the preceding control philosophies, however, response to changes in the shop floor status is quite slow and timely decision is sometimes impossible. Moreover, the control software becomes too large and it is almost impossible to modify the control software when the configuration of the shop floor changes. In agile manufacturing which emerged recently to cope with quick response and easy modifiability when unexpected changes occur, a new control policy is needed. CSCW[Computer Supported Cooperative Work] based shop floor control casts a different view on scheduling problems. Decisions are made locally when requested and useful information is scattered among agents for its efficient use. Adaptation is easy because agents are -'plug compatible or portable. In this paper, scheduling problems occurring under CSCW based shop floor control are identified and characterized. Traditional scheduling problems are reviewed from the CSCW viewpoint. All the control entities involved in the shop floor can be found and used to defined agents. With these entities and CSCW concept, possible scheduling problems are identified.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.1B
• /
• pp.11-17
• /
• 2003

In this paper, we presented the dynamic random access channel allocation method under the priority based scheduling policy in order to improve the system performance of HIPERLAN/2 standardized by ETSI According to the scheduling policy, AP scheduler primarily allocates the resource to the collision MT This scheduling policy bring about decreasing the transmission delay of collision MT Dynamic RCH(random access channel) allocation method decreases the collision probability by increasing the number of RCH slots in case of low traffic. While it increases the maximum throughput by increasing the number of the data transmission slots in case of high traffic Therefore dynamic allocation method of RCH slots decreases the scheduling delay and increases the throughput When we evaluate the performance of presented method based on standards, we saw that the presented method improve the performance of the MAC protocol in terms of throughput and transmission delay.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.2
• /
• pp.71-76
• /
• 2022

Round-robin is widely used for the scheduling of large-scale parallel workloads in the computing units of GPGPU. Round-robin is easy to implement by sequentially allocating tasks to each computing unit, but the load balance between computing units is not well achieved in multi-workload environments like cloud. In this paper, we propose a new thread block scheduling policy to resolve this situation. The proposed policy manages thread blocks generated by various GPGPU workloads with multiple queues based on their computation loads and tries to maximize the resource utilization of each computing unit by selecting a thread block from the queue that can maximally utilize the remaining resources, thereby inducing load balance between computing units. Through simulation experiments under various load environments, we show that the proposed policy improves the GPGPU performance by 24.8% on average compared to Round-robin.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.49-54
• /
• 2022

With the recent widespread adoption of general-purpose GPUs (GPGPUs) in cloud systems, maximizing the resource utilization through multitasking in GPGPU has become an important issue. In this article, we show that resource allocation based on the workload classification of computing-bound and memory-bound is not sufficient with respect to resource utilization, and present a new thread block scheduling policy for GPGPU that makes use of fine-grained resource utilizations of each workload. Unlike previous approaches, the proposed policy reduces scheduling overhead by separating profiling and scheduling, and maximizes resource utilizations by co-locating workloads with different bottleneck resources. Through simulations under various virtual machine scenarios, we show that the proposed policy improves the GPGPU throughput by 130.6% on average and up to 161.4%.

• Journal of applied mathematics & informatics
• /
• v.15 no.1_2
• /
• pp.511-525
• /
• 2004

we model communication and computer systems that process interactive and several and several types of background jobs. The scheduling policy in use is to share the processor among all interactive jobs and, at most, one background job of each type at a time according to the process sharing discipline. Background jobs of each type are served on a first-come-first-served basis. Such scheduling policy is called Processor Sharing with Background jobs (PSBJ). In fact, the PSBJ policy is commonly used on many communication and computer systems that allow interactive usage of the systems and process certain jobs in a background mode. In this paper, the stability conditions for the PSBJ policy are given and proved. Since an exact analysis of the policy seems to be very difficult, an approximate analytic model is proposed to obtain the average job sojourn times. The model requires the solution of a set of nonlinear equations, for which an iterative algorithm is given and its convergence is proved. Our results reveal that the model provides excellent estimates of average sojourn times for both interactive and background jobs with a few percent of errors in most of the cases considered.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.11
• /
• pp.9-14
• /
• 2008

In this paper, we propose a collision avoidance scheduling to increase the multiuser diversity gains in the adaptive orthogonal frequency division multiple access (OFDMA) system. The scheduling policy is based on the minimum collision criterion which investigates the differences of user channels. The paper includes the derivation of capacity expressions for the adaptive OFDMA system with the proposed scheduling. The analysis shows that the capacity of the system depends on the number of collisions between the selected users to be simultaneously served. Numerical results show that the proposed scheduling provides improved capacity performance over existing ones.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.2
• /
• pp.171-186
• /
• 1996

This paper comes from an observation that overemphasis on capacity utilization meausre, which is usual under capacity shortage, can seriously hurt the firm's profit and potential process improvement. We suggest a model that can be used in designing a coordination scheme for decentralized marketing and manufacturing activities. Using a price and time-sensitive demand and capacitated lotsizing model, we derive an effective communication medium betwen marketing and manufacturing. This Balance Indicator of process capacity and flexibility also implies that the increase in capacity availability and setup time reduction should be balanced by its market requirements. This is particularly important when a firm tries to improve its process capability by kaizen. Further, the model can be used to show the comparative performances of scheduling policies under capacity imbalnce. We show the shortening the scheduling cycle can improve the firm profit without changing the simple scheduling rule.