Mechanisms for monitoring optimization in cloud computing environments
| dc.contributor.advisor | Κυριαζής, Δημοσθένης | |
| dc.contributor.advisor | Kyriazis, Dimosthenis | |
| dc.contributor.author | Totow Tom-Ata, Jean - Didier | |
| dc.date.accessioned | 2021-01-22T06:37:13Z | |
| dc.date.available | 2021-01-22T06:37:13Z | |
| dc.date.issued | 2020-02 | |
| dc.identifier.uri | https://dione.lib.unipi.gr/xmlui/handle/unipi/13194 | |
| dc.identifier.uri | http://dx.doi.org/10.26267/unipi_dione/617 | |
| dc.format.extent | 86 | el |
| dc.language.iso | en | el |
| dc.publisher | Πανεπιστήμιο Πειραιώς | el |
| dc.rights | Αναφορά Δημιουργού 4.0 Διεθνές | * |
| dc.rights | Αναφορά Δημιουργού 4.0 Διεθνές | * |
| dc.rights | Αναφορά Δημιουργού 4.0 Διεθνές | * |
| dc.rights | Αναφορά Δημιουργού 4.0 Διεθνές | * |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | * |
| dc.title | Mechanisms for monitoring optimization in cloud computing environments | el |
| dc.title.alternative | Μηχανισμοί βελτιστοποίησης εποπτείας σε περιβάλλοντα υπολογιστικών νεφών | el |
| dc.type | Master Thesis | el |
| dc.contributor.department | Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών Συστημάτων | el |
| dc.description.abstractEN | In big data environment that delivers a complete pioneering stack, based on a frontrunner infrastructure management system that drives decisions according to data aspects, thus being fully scalable, runtime adaptable and high-performant to address the emerging needs of big data operations and data-intensive applications, the data-driven platform should collect and analyse/evaluate periodically metrics from different components involved in a specific application performance. We distinguish three groups of components involved in the environment performance: the infrastructure where applications are running (Kubernetes1, openshift2 etc …), data components (object storing systems, databases) and applications running. These components generate a huge amount of metrics which have to be collected, evaluated (quality of service) stored and exposed to a decision component in real-time and an ad-hoc mode. Metric could be memory usage, a cpu consumption, number of processes, application starting time etc. We can clearly understand that some metrics could be produced by a batch job and some others are produced periodically so the need of providing different mechanisms to collect and consume them. Building a monitoring engine implementing functionalities listed above introduces a considerable delay from the moment a metric is collected and the moment this metric is available for consumption due to all processing units in between. The bigger is the amount of measurements, the more information the platform can receive and better will be the decision. However, the amount of data is directly proportional to the delay related earlier. This delay affects the performance of the decision component since this last should catch events as soon as possible. In order to enable later analysis on metrics, the monitoring engine should provide methods for storing metrics. However, measurements are taken periodically from applications for being used for analysis and historical purpose. | el |
| dc.contributor.master | Πληροφοριακά Συστήματα και Υπηρεσίες | el |
| dc.subject.keyword | Monitoring | el |
| dc.subject.keyword | Cloud computing | el |
| dc.subject.keyword | Data-driven | el |
| dc.date.defense | 2020-02-17 |
Αρχεία σε αυτό το τεκμήριο
Αυτό το τεκμήριο εμφανίζεται στις ακόλουθες συλλογές
-
Τμήμα Ψηφιακών Συστημάτων
Department of Digital Systems
Εκτός από όπου διευκρινίζεται διαφορετικά, το τεκμήριο διανέμεται με την ακόλουθη άδεια:
Αναφορά Δημιουργού 4.0 Διεθνές
Αναφορά Δημιουργού 4.0 Διεθνές