東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Improving reliability through task r...

Nimer, Bayan Abdel Jalil.

FindBook

Google Book

Amazon

博客來

Improving reliability through task recomputation in embedded systems.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Improving reliability through task recomputation in embedded systems./
作者:	Nimer, Bayan Abdel Jalil.
面頁冊數:	58 p.
附註:	Source: Masters Abstracts International, Volume: 51-06.
Contained By:	Masters Abstracts International51-06(E).
標題:	Engineering, Computer. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1523004
ISBN:	9781303177026

Improving reliability through task recomputation in embedded systems.
Nimer, Bayan Abdel Jalil.

Improving reliability through task recomputation in embedded systems. - 58 p.

Source: Masters Abstracts International, Volume: 51-06.

Thesis (M.S.)--University of Houston-Clear Lake, 2013.

The demand for high performance and computational complexity, driven by the recent advances in technology such as high clock rate, shrinking transistor size and lower power consumption techniques, continue to increase in embedded systems leading to an increased soft error rate. In some applications, such as video, these errors are displayed as wrong colored bits and are not problematic; however, in safety critical applications these errors could be catastrophic when occurring during critical parts of an operation leading to a complete loss of functionality and sometimes jeopardizing human lives [48]. As a result, reliability issues are becoming limiting factors on application scalability and longterm survivability in the harsh conditions in which such systems are operated. As embedded systems already house large datasets, traditional techniques to improve reliability such as task redundancy can be very costly in terms of memory space consumption. Earlier research conducted on re-computation has successfully proved to decrease memory space consumption in embedded systems [28, 30, 31, 41, 50]. Consequently, the idea of utilizing ideal times to re-compute a task triggered the interest that these idle time frames of computational resources could be utilized to iteratively recompute tasks to increase reliability of overall design without incurring any additional execution latency, area or extra memory space. That is, all tasks are originally stored in memory and kept there until the end of their lifetimes and if a certain task fails to produce intended result, the recomputed task is used to produce the result and pass it to the next task that needs it without being stored in memory. To further investigate the role of re-computation on maximizing reliability we conducted experiments on random task graphs extracted using the TGFF tool [15], which generates graphs tailored to the areas of allocation and scheduling research in embedded systems. To gain further evidence on the effectiveness of the approach, its feasibility was evaluated on data intensive embedded benchmarks. The results that were observed in heterogeneous architectures and in HW/SW co-design show the effective role that task recomputation may play in increasing the robustness of embedded systems against transient errors. Therefore, incorporating recomputation during different abstraction levels of the design will expand capabilities to develop highly reliable embedded systems.

ISBN: 9781303177026Subjects--Topical Terms:

1669061
Engineering, Computer.

Improving reliability through task recomputation in embedded systems.
LDR:03341nam a2200289 4500 001 1960662
005 20140624205946.5
008 150210s2013 ||||||||||||||||| ||eng d
020 $a 9781303177026
035 $a (MiAaPQ)AAI1523004
035 $a AAI1523004
040 $a MiAaPQ $c MiAaPQ
100 1 $a Nimer, Bayan Abdel Jalil. $3 2096359
245 1 0 $a Improving reliability through task recomputation in embedded systems.
300 $a 58 p.
500 $a Source: Masters Abstracts International, Volume: 51-06.
500 $a Adviser: Hakduran Koc.
502 $a Thesis (M.S.)--University of Houston-Clear Lake, 2013.
520 $a The demand for high performance and computational complexity, driven by the recent advances in technology such as high clock rate, shrinking transistor size and lower power consumption techniques, continue to increase in embedded systems leading to an increased soft error rate. In some applications, such as video, these errors are displayed as wrong colored bits and are not problematic; however, in safety critical applications these errors could be catastrophic when occurring during critical parts of an operation leading to a complete loss of functionality and sometimes jeopardizing human lives [48]. As a result, reliability issues are becoming limiting factors on application scalability and longterm survivability in the harsh conditions in which such systems are operated. As embedded systems already house large datasets, traditional techniques to improve reliability such as task redundancy can be very costly in terms of memory space consumption. Earlier research conducted on re-computation has successfully proved to decrease memory space consumption in embedded systems [28, 30, 31, 41, 50]. Consequently, the idea of utilizing ideal times to re-compute a task triggered the interest that these idle time frames of computational resources could be utilized to iteratively recompute tasks to increase reliability of overall design without incurring any additional execution latency, area or extra memory space. That is, all tasks are originally stored in memory and kept there until the end of their lifetimes and if a certain task fails to produce intended result, the recomputed task is used to produce the result and pass it to the next task that needs it without being stored in memory. To further investigate the role of re-computation on maximizing reliability we conducted experiments on random task graphs extracted using the TGFF tool [15], which generates graphs tailored to the areas of allocation and scheduling research in embedded systems. To gain further evidence on the effectiveness of the approach, its feasibility was evaluated on data intensive embedded benchmarks. The results that were observed in heterogeneous architectures and in HW/SW co-design show the effective role that task recomputation may play in increasing the robustness of embedded systems against transient errors. Therefore, incorporating recomputation during different abstraction levels of the design will expand capabilities to develop highly reliable embedded systems.
590 $a School code: 1251.
650 4 $a Engineering, Computer. $3 1669061
650 4 $a Computer Science. $3 626642
650 4 $a Engineering, General. $3 1020744
690 $a 0464
690 $a 0984
690 $a 0537
710 2 $a University of Houston-Clear Lake. $3 1018602
773 0 $t Masters Abstracts International $g 51-06(E).
790 $a 1251
791 $a M.S.
792 $a 2013
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=1523004