• The KIPS Transactions:PartA
• /
• v.18A no.4
• /
• pp.143-150
• /
• 2011

This paper presents an EDF based scheduling algorithm for scheduling imprecise computation model where each task consists of mandatory part and optional part. Imprecise computation is useful to manage overload condition. In overload situation, some optional parts should be removed. The proposed DOP algorithm removes optional parts of earlier deadline tasks to enhance flexibly for newly arriving tasks. A simulation result shows that DOP has better performance than other algorithms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.5
• /
• pp.783-791
• /
• 2006

In e-Trade, Real-Time System(RTS) plays a very important role. Each task is set with limited time, and appointed regulations must be followed because it can be greatly damaged if it cannot be executed in limited time. In e-Trade, the scheduling possibility techniques generally use periodical tasks; however, it is necessary to study more stable prediction scheduling possibility algorithm by using other task timing conditions and non-periodical task scheduling tasks. This study proposed an algorithm to increase the prediction possibility using individual task utilization rate, and presented scheduling possibility conditions using existing whole task utilization rate and the proposed algorithm.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.4
• /
• pp.443-454
• /
• 1999

최소 여유시간 우선(Least-Laxity-First)스케쥴링 알고리즘은 여유시간이 작은 태스트가 높은 우선순위를 갖는 동적 우선순위 스케줄링 방법으로서 단일프로세서 시스템에서 최적임이 증명되었다. 그러나, 이 알고리즘은 최소 여유시간을 가진 태스크가 여러 개 존재하여 여유시간 충돌이 발생한 경우 이 태스크들 간에 빈번한 문맥교환이 발생하게 되는 문제점이 있어 실용적이지 못하다. 본 논문에서 제한하는 최소 여유시간 우선 기반의 최소선점을 갖는 스케줄링 알고리즘(Least-Laxity-First with Minimum Preemption 또는 LLF/MP)은 여유시간이 충돌했을 때에 문맥교환을 최소화함으로써 최소 여유시간 우선 스케줄링 알고리즘의 단점을 해결하였다. LLF/MP 스케줄링 알고리듬은 불필요한 문맥교환을 줄임으로써 시스템 오버헤드로 인한 시스템의 성능 저하를 방지할 수 있고 보다 많은 시스템 자원을 예측 불가능한 비주기적 태스크에게 할당할수 있다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.2 s.308
• /
• pp.12-20
• /
• 2006

In scheduling of real-time tasks, the required hardware performance for a given set of tasks is determined based on the worst case execution time. For soft real-time tasks as multimedia applications, a lower performance hardware can service the tasks. Since the execution time of a task can vary in time, we can reclaim the slacks of early completed tasks for those of longer than average execution times. Then, the average ratio of deadline-miss can be lowered. This paper presents an algorithm, Aggressive Slack Reclamation (ASR), that tasks share slacks aggressively. A simulation result shows that ASR enhances the deadline-miss ratio and number of context switches than previous results.

• The KIPS Transactions:PartA
• /
• v.16A no.6
• /
• pp.437-442
• /
• 2009

For multiprocessor systems, Earliest Deadline First (EDF) based on deadline and Least Laxity First (LLF) based on laxity are not suitable for practical environment since EDF has low schedulability and LLF has high context switching overhead. As a combining of EDF and LLF to improve the performance, Earliest Deadline Zero Laxity (EDZL) was proposed. EDZL is basically the same as EDF. But if the laxity of a task becomes zero, its priority is promoted to the highest level. In this paper, we propose Least Laxity Zero Laxity (LLZL) which is based on LLF. But context switching is allowed only if the laxity of a task on rady queue becomes zero. Simulation results show that LLZL has high schedulability approaching to LLF and low context switching overhead similar to EDF. In comparison with EDZL, LLZL has better performance in both of schedulability and context switching overhead.

• Journal of Korea Multimedia Society
• /
• v.14 no.11
• /
• pp.1478-1490
• /
• 2011

Recently, multi-core processors have been drawing significant interest from the embedded systems research and industry communities due mainly to their potential for achieving high performance and fault-tolerance at low cost in such products as automobiles and cell phones. To process multimedia data, a scheduling algorithm is required to meet timing constraints of periodic tasks in the system. Though Pfair scheduling algorithm can meet all the timing constraints while achieving 100% utilization on multi-core based system theoretically, however, the algorithm incurs high scheduling overheads including frequent core migrations and system-wide synchronizations. To mitigate the problems, we propose a real-time scheduling algorithm for multi-core based system so that system-wide scheduling is performed only when it is absolutely necessary. Otherwise the proposed algorithm performs scheduling within each core independently. The experimental results by extensive simulations show that the proposed algorithm dramatically reduces the scheduling overheads up to as negligible one when the utilization is under 80%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.1
• /
• pp.88-95
• /
• 2004

The previous disk scheduling schemes for best-effort applications do not guarantee the real-time requirement of multimedia objects and the real-time disk scheduling schemes do not satisfy throughput of multimedia server. So, this paper propose a two-step disk scheduling scheme to satisfy the requirement of best-effort as well as soft real-time applications. This scheme is based on the round robin algorithm that imposes different weights on the best-effort task and the real-time one. The experiment results on the Linux kernel have shown that both best-effort tasks and real-time tasks could get fair service.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.8B
• /
• pp.509-520
• /
• 2007

In this paper, we propose a real-time sensor node platform that guarantees the real-time scheduling of periodic and aperiodic tasks through a multitask-based software decomposition technique. Since existing sensor networking operation systems available in literature are not capable of supporting the real-time scheduling of periodic and aperiodic tasks, the preemption of aperiodic task with high priority can block periodic tasks, and so periodic tasks are likely to miss their deadlines. This paper presents a comprehensive evaluation of how to structure periodic or aperiodic task decomposition in real-time sensor-networking platforms as regard to guaranteeing the deadlines of all the periodic tasks and aiming to providing aperiodic tasks with average good response time. A case study based on real system experiments is conducted to illustrate the application and efficiency of the multitask-based dynamic component execution environment in the sensor node equipped with a low-power 8-bit microcontroller, an IEEE802.15.4 compliant 2.4GHz RF transceiver, and several sensors. It shows that our periodic and aperiodic task decomposition technique yields efficient performance in terms of three significant, objective goals: deadline miss ratio of periodic tasks, average response time of aperiodic tasks, and processor utilization of periodic and aperiodic tasks.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.12
• /
• pp.736-748
• /
• 2003

Dynamic voltage scaling(DVS), which adjusts the clock speed and supply voltage dynamically, is an effective technique in reducing the energy consumption of embedded real-time systems. The energy efficiency of a DVS algorithm largely depends on the performance of the slack estimation method used in it. In this paper, we propose novel DVS algorithms for periodic hard real-time tasks based on an improved slack estimation algorithm. Unlike the existing techniques, the proposed method can be applied to most priority-driven scheduling policies. Especially, we apply the proposed slack estimation method to EDF and RM scheduling policies. The experimental results show that the DVS algorithms using the proposed slack estimation method reduce the energy consumption by 20∼40 % over the existing DVS algorithms.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.11a
• /
• pp.329-334
• /
• 2007

감시정찰 센서네트워크의 모든 센서노드 및 싱크노드들은 한정된 자원과 저사양의 하드웨어로 동작하며, 각 침입탐지 센서들이 수집한 상황 데이터를 신뢰성 있게 전송할 수 있어야 한다. 본 초소형 내장소프트웨어는 이러한 감시정찰 센서네트워크의 특성에 맞게 설계되어 센서 및 싱크노드에 탑재될 수 있는 소프트웨어로서, 센서 OS 커널, 센서미들웨어, 보안커널로 구성된다. 센서 OS는 Multithread 기반으로 실시간, 비실시간 태스크를 위한 각기 다른 스케줄링 방식을 제공하며 지연된 인터럽트 처리 기능, 주기적 태스킹 기능과 효율적 에너지 관리 기능을 제공하여 센서 네트워크에 특화된 어플리케이션 개발을 용이하게끔 한다. 또한 센서미들웨어는 OS 커널과 어플리케이션 사이에 존재하여 위치인식, 시간동기, 네트워크 관리, 원격 업데이트 기능 등 어플리케이션에서 공통적으로 요구하는 필수 기능들을 제공한다.