DocumentCode :
121047
Title :
An Advance Reservation Mechanism to Enhance Throughput in an Opportunistic High Performance Computing Environment
Author :
Gomes, Eliza ; Ribeiro Dantas, Mario Antonio
Author_Institution :
Dept. of Inf. & Statistic (INE), Fed. Univ. of Santa Catarina (UFSC), Florianopolis, Brazil
fYear :
2014
fDate :
21-23 Aug. 2014
Firstpage :
221
Lastpage :
228
Abstract :
Opportunistic high performance computing environments have been widely used, especially by the academic community owing to the possibility of obtaining high processing power with low cost, considering that this type of distributed high performance environments uses idle computing resources of personal machines. However, excessive requests of resources allocation can be recurrent and can both degrade the performance of the system and make the process of resource request harder for the user, who might have to repeat this process until resources for execution are available. Therefore, this paper proposes the use of an advance reservation mechanism in an opportunistic environment. It aims to improve the use flow of opportunistic resources in order to offer the possibility of resources allocation for a long period of time and not only at the time of request. Case studies were carried out to illustrate the behavior of an opportunistic environment using the approach proposed here, and results showed the efficiency and validity of our proposal.
Keywords :
parallel processing; resource allocation; advance reservation mechanism; distributed high performance environments; opportunistic high performance computing environment; opportunistic resources flow; personal machines; resources allocation; throughput enhancement; Availability; Computer architecture; Databases; High performance computing; Laboratories; Proposals; Resource management; Advance Reservation of Resource; Opportunistic high performance computing environment; Resource Allocation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Network Computing and Applications (NCA), 2014 IEEE 13th International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
978-1-4799-5392-9
Type :
conf
DOI :
10.1109/NCA.2014.40
Filename :
6924231
Link To Document :
بازگشت