Mitigation of performance variability induced by Checkpoint-Restart using DVFS

Κοκόλης, Απόστολος; Kokolis, Apostolos

dc.contributor.author	Κοκόλης, Απόστολος	el
dc.contributor.author	Kokolis, Apostolos	en
dc.date.accessioned	2015-09-30T09:01:03Z
dc.date.available	2015-09-30T09:01:03Z
dc.date.issued	2015-09-30
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/41344
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.11434
dc.rights	Default License
dc.subject	Αξιοπιστία	el
dc.subject	Reliability	el
dc.subject	Διαθεσιμότητα	el
dc.subject	Λειτουργικότητα	el
dc.subject	Δυναμική μεταβολή τάσης και συχνότητας	el
dc.subject	Επιτάχυνση εκτέλεσης	el
dc.subject	Availability	en
dc.subject	Checkpoint/restart	en
dc.subject	Dynamic voltage and frequency scaling	en
dc.subject	Execution sprinting	en
dc.title	Mitigation of performance variability induced by Checkpoint-Restart using DVFS	en
heal.type	bachelorThesis
heal.classification	Πληροφορική και υπολογιστές	el
heal.language	el
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2015-07-28
heal.abstract	As performance enhancement is accompanied by the aggressive integration of many-cores to a single chip and technology nodes approach deca-nanometer dimensions, the system's failure rate is becoming signi cant. Inevitably, computer systems must tolerate such failures. Both hardware and software methods are available enabling fault-tolerance to the systems. The Checkpoint/Restart technique provides reliability to the execution of an application. However, Checkpoint/Restart introduce an additional time overhead in order to achieve the fault-tolerance of the execution, that leads to performance variability. The scope of this thesis is to enhance a runtime manager, Depman, that orchestrates an application level Checkpoint/Restart technique so that such time overheads are absorbed, achieving performance predictability and reliability on the y, by using Dynamic Voltage and Frequency Scaling (DVFS). A closed-loop implementation controlling the clock frequency is proposed, that quanti es the time overheads induced by the checkpoint restart process and adjusts the frequency levels of the CPU so that execution time converges to the normal. Depman was also modi ed to extend its portability to other platforms and applications and was tested using the self fault injection module to both the Intel's Single-Chip Cloud Computer (SCC) and an x86 general computing platform, evaluating both the execution time and energy consumption of our scheme.	en
heal.advisorName	Σούντρης, Δημήτριος	el
heal.committeeMemberName	Πεκμεστζή, Κιαμάλ	el
heal.committeeMemberName	Κοζύρης, Νεκτάριος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών. Τομέας Τεχνολογίας Πληροφορικής και Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	83 σ.
heal.fullTextAvailability	true