• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.199-201
• /
• 2012

디스크나 이더넷과 같은 I/O 장치로부터 발생하는 I/O 트래픽은, 여러 개의 노드를 가진 NUMA 시스템의 공유 LLC에 캐시 오염을 일으켜 캐시 라인이 재사용되는 것을 방해한다. 이러한 태스크는 캐시를 효율적으로 이용할 수 있는 메모리 집중적인 태스크들과 따로 분리하여 다룰 필요가 있다. 본 논문에서는 이러한 캐시 오염을 발생시키는 태스크들을 해당 태스크의 I/O 트래픽을 이용하여 실시간으로 감시하고 분류하는 기법을 제안한다. 또한 대량의 I/O 트래픽을 일으키는 태스크의 특성을 알아본다. 이를 통해, NUMA 시스템 환경에서 각 노드의 공유 LLC를 보다 효율적으로 사용할 수 있는 운영체제 스케줄링 기법을 연구하기 위한 토대를 마련하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.457-460
• /
• 2013

서버 및 PC 시장에서의 멀티코어 프로세서의 강세는 임베디드 기기에도 이어지고 있으며 최근 이기종 멀티코어 프로세서를 탑재한 임베디드 제품들도 출시되고 있다. 태스크 스케줄러 관점에서 멀티코어 프로세서는 태스크들이 효과적으로 스케줄링 될 수 있도록 코어를 선택하고 태스크의 이주를 통해 다른 코어들과의 로드를 유지해야 한다. 그러나 현재 임베디드 기기의 태스크 스케줄러는 모든 코어에 동일한 정책을 적용함으로써 태스크의 특징에 따른 효과적인 자원관리를 못하고 있다. 본 논문에서는 코어별로 스케줄링 정책을 관리하는 기법을 적용함으로써 태스크의 특징에 따른 코어의 활용을 높일 수 있는 방안을 제시한다.

• The Journal of the Korea Contents Association
• /
• v.21 no.6
• /
• pp.12-18
• /
• 2021

A proportional share scheduler allocates CPU time to tasks and determines which task will be dispatched according to their priorities. In this paper, we investigate the correlation between the characteristics of applications and task priorities in the Linux Completely Fair Scheduler(CFS), which is one of the representative proportional share schedulers. We also propose a method of controlling the granularity of priority assignments according to the characteristics of applications. We implemented the proposed technique in a Linux system and confirmed the meaningful experimental results.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.147-150
• /
• 2022

최근 들어 다양한 제약 조건이 있는 스마트 시티나 스마트 팩토리와 같은 도메인들 내에서 태스크들을 효과적으로 처리하기 위해서 MEC 기술이 많이 사용되고 있다. 그러나 이러한 도메인에서 발생하는 복잡하고 동적인 시나리오는 기존의 휴리스틱이나 메타 휴리스틱 기법을 이용하여 해결하기엔 계산 복잡도가 증가하는 문제점을 가지고 있다. 따라서 최근 들어 이러한 문제점을 해결하기 위한 방법 중 하나로 강화학습과 딥러닝이 결합된 DRL 기법이 주목을 받고 있다. 본 연구는 스마트 팩토리 환경에서 종속성을 가진 태스크들이 실행시간과 태스크가 처리되는 MEC 서버들의 로드 표준편차를 최소화하는 태스크 스케줄링 기법을 제안한다. 모의실험을 통하여 제안 기법은 태스크가 증가하는 동적인 환경에서도 좋은 성능을 보임을 증명하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2235-2246
• /
• 1997

This paper presents an effective heuristic task scheduling algorithm for multiprocessor systems. To execute task scheduling effectively which is defined as an allocation of m's tasks onto n's processors(m > n), several problems almost at NP-hard should be cleaned up. The purpose of the task scheduling obtains the minimum execution time by mapping the tasks on a system topology or reduces the total execution time to give a minimum system topology. In order to solve this problem, in this paper, the task scheduling is done by redefining a task graph to a reconstructed task graph (RTG). An RTG is obtained by merging or copying nodes to equal the number of nodes on each level of the task graph to the number of processors of the system topology and then directly scheduled to the system topology. This method obtains a fast scheduling time and a simple scheduling method, and near-optimal execution time without executing steps such as the refinement step and the duplication step after the task scheduling.

• Journal of Korea Multimedia Society
• /
• v.12 no.5
• /
• pp.617-632
• /
• 2009

In this paper we propose a technique of implementing the reconfigurable network protocol stack that allows for partitioning network protocol functions into software and hardware tasks on a SoC (System on Chip) platform. Additionally, we present a method that guarantees the deadline of both an individual task and messages exchanging among tasks in order to meet the deadline of real-time multimedia and networking services. The proposed real-time message exchange method guarantees the deadline of messages generated by multimedia services that are required to meet the real-time properties of multimedia applications. After implementing the networking functions of TCP/IP protocol suite into hardware and software tasks, we verify and validate their performance on the SoC platform. Experimental results indicate that the proposed technique improves the performance of TCP/IP protocol suit as well as application service satisfaction in application-specific real-time.

• KIISE Transactions on Computing Practices
• /
• v.23 no.7
• /
• pp.452-457
• /
• 2017

Linux is a general-purpose operating system that supports several schedulers, allowing different schedulers to coexist. In addition, Linux uses the Control Group (cgroup) to reserve CPU resources for task groups that follow the real-time (SCHED_FIFO, SCHED_RR) and non-real-time (SCHED_NORMAL) scheduler policies, except for the deadline scheduler (SCHED_DEADLINE). The cgroup performs the schedulability analysis to guarantee the reserved CPU resource as much as possible. However, current implementation of the schedulability analysis does not distinguish between deadline tasks and real-time tasks. Therefore, if these deadline tasks and real-time task groups coexist, there is a case where the resource reservation for the real-time task group is rejected. In this paper, we analyze the problems in the schedulability analysis for real-time task groups of Linux cgroups and propose patches to solve these problems.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.5
• /
• pp.506-517
• /
• 2000

In distributed real-time system, a task activated by the completion of its preceding task can be modeled as a periodic task with activation jitter. An activation jitter of a task is defined asthe difference between the worst case and the best case response time of its preceding task. Becausethe existing approaches assume that the best case response time is much smaller than the actual one,the activation jitter and the worst case response time of lower priority tasks are overestimated. Thispaper proposes a new analysis technique to calculate the best case response time more precisely andto reduce the activation jitter bounds. The proposed technique obtains the best case response time byconsidering the relative phase between tasks. The precise analysis of the activation jitters can reducethe worst case response time of other tasks and increase the schedulability. The simulation resultsshow that the proposed analysis technique improves the accuracy of the best case and the worst caseresponse time up to 40% and 6%, respectively.

• The KIPS Transactions:PartA
• /
• v.10A no.3
• /
• pp.181-188
• /
• 2003

We propose a heuristic on-line scheduling algorithm for the IRIS (Increasing Reward with Increasing Service) tasks, which has low computation complexity and produces total reward approximated to that of previous on-line optimal algorithms. The previous on-line optimal algorithms for IRIS tasks perform scheduling on all tasks in a system to maximize total reward. Therefore, the complexities of these algorithms are too high to apply them to practical systems handling many tasks. The proposed algorithm doesn´t perform scheduling on all tasks in a system, but on (constant) W´s tasks selected by a predefined task selection policy. The proposed algorithm is based on task selection policies that define how to select tasks to be scheduled. We suggest two simple and intuitive selection policies and a generalized selection policy that integrates previous two selection policies. By narrowing down scheduling scope to only W´s selected tasks, the computation complexity of proposed algorithm can be reduced to O(Wn). However, simulation results for various cases show that it is closed to O(W) on the average.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.11
• /
• pp.573-580
• /
• 2001

When tasks access shared resources in real-time systems, the unbounded priority inversion may occur. In such cases it is impossible to guarantee the schedulability of real-time tasks. Several resource access protocols have been proposed to bound the duration of priority inversion and sufficient conditions are given to guarantee the schedulability of periodic task set. In this paper, we show an improved sufficient condition for schedulability when the dynamic priority ceiling protocol is used. Our approach exploits the fact that a lower priority task can continue to execute as far as the higher priority tasks do not miss their deadlines. This permitting execution time of the higher priority tasks for a lower priority task can be excluded from the worst-case blocking time of the higher priority tasks. Since the worst-case blocking time of tasks can be reduced, the sufficient condition for schedulability of dynamic priority ceiling protocol becomes further tight.