• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.1
• /
• pp.87-95
• /
• 2024

This study is on modulo scheduling algorithms for multicore processor in machine learning applications. Machine learning algorithms are designed to perform a large amount of operations such as vectors and matrices in order to quickly process large amounts of data stream. To support such large amounts of computations, processor architectures to support applications such as artificial intelligence, neural networks, and machine learning are designed in the form of parallel processing such as multicore. To effectively utilize these multi-core hardware resources, various compiler techniques are being used and studied. In this study, among these compiler techniques, we analyzed the modular scheduler, which is especially important in one core's computation pipeline. This paper looked at and compared the iterative modular scheduler and the swing modular scheduler, which are the most widely used and studied. As a result, both schedulers provided similar performance results, and when measuring register pressure as an indicator, it was confirmed that the swing modulo scheduler provided slightly better performance. In this study, a technique that divides recurrence edge is proposed to improve the minimum initiation interval of the modulo schedulers.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.7B
• /
• pp.433-439
• /
• 2005

In this paper, we propose a Flow Based Waiting Time Priority (FB-WTP) scheduler that supports the proportional delay differentiated services between classes. Existing scheduling algorithms utilize the configuration of scheduler, which is operated as class unit, applied to wired network. However, FB-WTP scheduler is operated as flow unit and can take advantage of multi-flow diversity effect in time-varying channel state environment. As a result, FB-WTP improves the average queueing delay on each class as well as supporting the average queueing delay ratio between classes. It also solves the HOL packet blocking problem implicitly. Simulation results show that FB-WTP scheduler has better system queueing delay performance than Look-ahead Waiting Time Priority (LWTP) scheduler and supports the larger queueing delay ratio between classes that network operator set.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.8
• /
• pp.1508-1516
• /
• 2017

Multipath TCP (MPTCP) is a transport layer protocol that supports multipath transmission. If a bufferbloat occurs in one of the subflows of MPTCP, HoL blocking occurs at the receiver due to the difference in packet arrival time among paths. In MPTCP, HoL blocking degrades not only the performance of the path where bufferbloat occurs, but also the performance of other paths. In this paper, we propose a bandwidth measurement based scheduler to solve this problem. Bandwidth measurement based scheduler is designed to measure the bandwidth of each subflow and to perform packet scheduling based on it. In order to verify the proposed scheduler, we implemented the proposed scheduler in the Linux kernel and constructed a testbed in which bufferbloat occurs. Experimental results show that the proposed scheduler has better performance than the legacy MPTCP in bufferbloat environment.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.1
• /
• pp.38-46
• /
• 2011

The OPRoS (Open Platform for Robotic Service) framework provides uniform operating environment for service robots. As an OPRoS-based service robot has to support real-time as well as non-real-time applications, application of Windows NT kernel based operating system can be restrictive. On the other hand, various benefits such as rich library and device support and abundant developer pool can be enjoyed when service robots are built on Windows NT. The paper presents a user-mode component scheduler of OPRoS, which can provide near real-time scheduling service on Windows NT based on the restricted real-time features of Windows NT kernel. The component scheduler thread with the highest real-time priority in Windows NT system acquires CPU control. And then the component scheduler suspends and resumes each periodic component executors based on its priority and precedence dependency so that the component executors are scheduled in the preemptive manner. We show experiment analysis on the performance limitations of the proposed scheduling technique. The analysis and experimental results show that the proposed scheduler guarantees highly reliable timing down to the resolution of 10ms.

• The KIPS Transactions:PartA
• /
• v.12A no.5 s.95
• /
• pp.375-384
• /
• 2005

General purpose operating systems are Increasingly being used for serving time-sensitive applications. These applications require soft real-time characteristics from the kernel and from other system-level services. In this paper, we explore various operating systems techniques needed to support time-sensitive applications and describe the design of MUSMA(Multi-level Scheduler for Multimedia Application). MUSMA is a framework that combination of user-level top scheduler and kernel-level bottom scheduler. We develope MUSMA in linux environment and it's performance is evaluated. Experiment result shows that it is possible to satisfy the constraints of multimedia in a general purpose operating system without significantly compromising the performance of non-realtime applications.

• Journal of IKEEE
• /
• v.21 no.1
• /
• pp.77-80
• /
• 2017

In this paper, we proposed a way to improve performance of GP-GPU by deletion of superscalar issue from its original form. At first, we simplified the structure of stream processor in order to eliminate superscalar issue. Under this condition, preservation of hardware size and increasing of thread number were followed by functional improvement of GP-GPU. As the number of thread was getting larger, we proposed the new model of warp scheduler which adjusts the group of thread. This superscalar issue-deleted warp scheduler transferred the instructions to warp which was activated by Round Robin Scheduling. Performance comparison was conducted by Gaussian filtering and the results indicated that our newly designed GP-GPU showing 7.89 times better in its performance than original one.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.8
• /
• pp.1827-1834
• /
• 2013

The existing PC-based desktop environment is being changed to a server-based virtual desktop environment due to various advantages such as security, mobility, and upgrade cost reduction. In this paper, a virtual computing lab service system which is applicable to the existing computer lab is designed and implemented using both an open source-based cloud computing platform and hypervisor. In addition, a node power scheduler is proposed in order to reduce power consumption in a server farm. The experimental results show that the power scheduler reduces power consumption considerably over the server farm without the power scheduler.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.7
• /
• pp.788-812
• /
• 1999

본 논문에서는 범용 워크스테이션 환경하에서 수행되는 멀티미디어 응용프로그램(application)을 지원하기 위한 비례지분 방식의 CPU 스케줄러를 제시한다. 이러한 목적을 위하여 일반적 태스크의 지원을 위해 설계된 스트라이드 스케줄러를 확장한다. 멀티미디어 응용프로그램의 시간 요구사항을 명시하기 위하여 새로운 스케줄링 파라미터들을 도입한다. 비율조정기를 도입한 결과 스케줄링의 정확도의 오차는 O(1)로 감소하였다. 별도의 태스크 그룹을 설정하여 상대적 지분과 절대적 지분을 부여했다. 모의실험을 사용하여 스케줄러의 성능을 평가하였다. 그 결과, 제안된 스케줄러는 증가된 정확도와 적응성 및 유연성을 가짐을 알 수 있었다. Abstract This paper presents a proportional-share CPU scheduler which can support multimedia applications in a general-purpose workstation environment. For this purpose, we have extended the stride scheduler which is designed originally for conventional tasks. New scheduling parameters are introduced to specify timing requirements of multimedia applications. Through the use of the rate regulator, the accuracy error of the scheduling is reduced to O(1). Separate task groups are proposed to represent both relative shares and absolute shares. The proposed scheduler is evaluated using a simulation study. The results show that the proposed scheduler achieves improved accuracy and adaptability as well as flexibility.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.5
• /
• pp.236-239
• /
• 2006

A wireless MAC scheduler that provides a high level of quality-of-service (QoS) for video-on-demand (VOD) applications while achieving a reasonable level of system throughput is proposed. The proposed scheduler considers both channel qualities of mobiles and the urgency of real-time packets coming from VOD applications in a cross-layer approach between application and MAC layers.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9B
• /
• pp.775-785
• /
• 2008

We identify the problem of the current IEEE 802. lie HCCA (Hybrid Coordination Function Controlled Channel Access) scheduler and its numerous variations, that the queue information cannot be notified to the Hybrid Coordinator (HC) timely, therefore the uplink delay lengthens unnecessarily. We suggests a simple solution and a couple of implementation practices, namely the Adaptive Scheduler with RTS/CTS (ASR) and Adaptive Scheduler with Data/Ack (ASD). They are both further elaborated to emulate the Deficit Round Robin (DRB) scheduler. They are finally compared with existing exemplary schedulers through simulations, and shown to perform well.