國立高雄大學圖資館 |

Language: English

Back

利用預測機制積極回收混合性任務未使用時間之排程演算法 = Aggress...

國立高雄大學資訊工程學系碩士班

利用預測機制積極回收混合性任務未使用時間之排程演算法 = Aggressive Reclaim with Prediction Algorithm

Record Type:	Language materials, printed : monographic
Paralel Title:	Aggressive Reclaim with Prediction Algorithm
Author:	黃丞毅,
Secondary Intellectual Responsibility:	國立高雄大學
Place of Publication:	高雄市
Published:	國立高雄大學;
Year of Publication:	2015[民104]
Description:	44葉圖，表格 : 30公分;
Subject:	混合性任務
Subject:	Mixed Task Set
Online resource:	https://hdl.handle.net/11296/f833tb
Notes:	107年11月1日公開
Notes:	參考書目：葉36-37
Summary:	在許多系統中會需要週期性的執行某些任務,與非週期性地不定時執行使用者查詢或是執行要求,如何以節約能源的方式執行系統中的任務,且能夠儘早完成非週期性的需求,是一個重要的研究議題。本篇論文主要為研究即時系統中混合性任務集合的排程問題,當演算法排程混合性任務集合時必須考慮系統能耗和非週期性工作集合的平均反應時間。在論文中,我們提出利用預測機制積極回收混合性任務未使用時間之排程演算法(Aggressive Reclaim with Prediction Algorithm:ARPA),本方法利用分析過去非週期性工作的抵達資料,在回收TB Server預留但未被使用的時間時,預測未來下一個非週期性工作的抵達時間,減少因過多時間的回收導致非週期性工作抵達時,截限時間會被設定較大而造成反應時間變差的情況。最後我們以實驗的方式,證實我們所提出的演算法,從實驗結果可以觀察到所提出的ARPA比按照比例積極回收機制(Ratio-based Aggressive Reclaim Algorithm:RARA)有更好的效能。 In many systems, some tasks will execute periodically and aperiodic user requests such as searching and execution commands are issued. It is an important research issue to execute tasks with considering simultaneously energy efficiency and finish aperiodic jobs as soon as possible. The objective of this paper is to study the scheduling problem of mixed task sets in real-time systems. The proposed scheduling algorithm for a mixed task set must minimize the energy consumption of the overall system, meet the timing constraints of periodic tasks, and minimize the average response time of aperiodic jobs. We propose the Aggressive Reclaim with Prediction Algorithm (ARPA), which analyzes historical arrival patterns of aperiodic jobs to predict the arrival time of the next aperiodic job. With prediction, execution time reserved for the TB Server but not used by aperiodic jobs can not be reclaimed too much and do not affect the deadline settings of aperiodic jobs. If the deadlines of virtual jobs are larger than arrival times of the next aperiodic jobs, the deadlines of aperiodic jobs become larger. Then the aperiodic jobs will have lager response times. A series of experiments were conducted to evaluate the proposed algorithm. The experimental results demonstrate that the performance of the proposed ARPA is better than Ratio-based Aggressive Reclaim Algorithm (RARA).

利用預測機制積極回收混合性任務未使用時間之排程演算法 = Aggressive Reclaim with Prediction Algorithm
黃, 丞毅

利用預測機制積極回收混合性任務未使用時間之排程演算法 = Aggressive Reclaim with Prediction Algorithm / 黃丞毅撰 - 高雄市 : 國立高雄大學, 2015[民104]. - 44葉 ; 圖，表格 ; 30公分.
107年11月1日公開參考書目：葉36-37.
混合性任務Mixed Task Set

利用預測機制積極回收混合性任務未使用時間之排程演算法 = Aggressive Reclaim with Prediction Algorithm
LDR:03613nam0a2200277 450 001 458396
005 20190102102605.0
010 0 $b 精裝
010 0 $b 平裝
100 $a 20151007d2015 k y0chiy50 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 利用預測機制積極回收混合性任務未使用時間之排程演算法 $d Aggressive Reclaim with Prediction Algorithm $z eng $f 黃丞毅撰
210 $a 高雄市 $c 國立高雄大學 $d 2015[民104]
215 0 $a 44葉 $c 圖，表格 $d 30公分
300 $a 107年11月1日公開
300 $a 參考書目：葉36-37
314 $a 指導教授：郭錦福教授
328 $a 碩士論文--國立高雄大學資訊工程學系碩士班
330 $a 在許多系統中會需要週期性的執行某些任務,與非週期性地不定時執行使用者查詢或是執行要求,如何以節約能源的方式執行系統中的任務,且能夠儘早完成非週期性的需求,是一個重要的研究議題。本篇論文主要為研究即時系統中混合性任務集合的排程問題,當演算法排程混合性任務集合時必須考慮系統能耗和非週期性工作集合的平均反應時間。在論文中,我們提出利用預測機制積極回收混合性任務未使用時間之排程演算法(Aggressive Reclaim with Prediction Algorithm:ARPA),本方法利用分析過去非週期性工作的抵達資料,在回收TB Server預留但未被使用的時間時,預測未來下一個非週期性工作的抵達時間,減少因過多時間的回收導致非週期性工作抵達時,截限時間會被設定較大而造成反應時間變差的情況。最後我們以實驗的方式,證實我們所提出的演算法,從實驗結果可以觀察到所提出的ARPA比按照比例積極回收機制(Ratio-based Aggressive Reclaim Algorithm:RARA)有更好的效能。 In many systems, some tasks will execute periodically and aperiodic user requests such as searching and execution commands are issued. It is an important research issue to execute tasks with considering simultaneously energy efficiency and finish aperiodic jobs as soon as possible. The objective of this paper is to study the scheduling problem of mixed task sets in real-time systems. The proposed scheduling algorithm for a mixed task set must minimize the energy consumption of the overall system, meet the timing constraints of periodic tasks, and minimize the average response time of aperiodic jobs. We propose the Aggressive Reclaim with Prediction Algorithm (ARPA), which analyzes historical arrival patterns of aperiodic jobs to predict the arrival time of the next aperiodic job. With prediction, execution time reserved for the TB Server but not used by aperiodic jobs can not be reclaimed too much and do not affect the deadline settings of aperiodic jobs. If the deadlines of virtual jobs are larger than arrival times of the next aperiodic jobs, the deadlines of aperiodic jobs become larger. Then the aperiodic jobs will have lager response times. A series of experiments were conducted to evaluate the proposed algorithm. The experimental results demonstrate that the performance of the proposed ARPA is better than Ratio-based Aggressive Reclaim Algorithm (RARA).
510 1 $a Aggressive Reclaim with Prediction Algorithm $z eng
610 # 0 $a 混合性任務 $a 系統能耗 $a 非週期性工作平均反應時間 $a 未使用執行時間
610 # 1 $a Mixed Task Set $a Energy Consumption $a Response Time $a Slack Time
681 $a 008M/0019 $b 464103 4410.1 $v 2007年版
700 1 $a 黃 $b 丞毅 $4 撰 $3 709640
712 0 2 $a 國立高雄大學 $b 資訊工程學系碩士班 $3 353878
801 0 $a tw $b NUK $c 20181115 $g CCR
856 7 # $u https://hdl.handle.net/11296/f833tb $z 電子資源 $2 http