dc.contributor.author |
Ηλιακοπούλου, Νικολέτα-Μαρκέλα
|
el |
dc.contributor.author |
Iliakopoulou, Nikoleta-Markela
|
en |
dc.date.accessioned |
2022-11-30T10:57:56Z |
|
dc.date.available |
2022-11-30T10:57:56Z |
|
dc.identifier.uri |
https://dspace.lib.ntua.gr/xmlui/handle/123456789/56312 |
|
dc.identifier.uri |
http://dx.doi.org/10.26240/heal.ntua.24010 |
|
dc.rights |
Default License |
|
dc.subject |
Virtualization |
en |
dc.subject |
Onboard data processing |
en |
dc.subject |
Raspberry Pi |
el |
dc.subject |
Εικονικοποίηση |
el |
dc.subject |
Jailhouse Hypervisor |
el |
dc.subject |
Real-time performance |
en |
dc.subject |
Isolation |
en |
dc.title |
Virtualization Techniques on Embedded
Systems with further application on Satellites |
en |
dc.contributor.department |
Microprocessors and Digital Systems Laboratory |
el |
heal.type |
bachelorThesis |
|
heal.classification |
Computer Science |
en |
heal.language |
el |
|
heal.language |
en |
|
heal.access |
free |
|
heal.recordProvider |
ntua |
el |
heal.publicationDate |
2022-07-12 |
|
heal.abstract |
Recent growth in the space industry has drawn attention to Satellite as a Service (SaaS). The primary goal of SaaS is to maximize the use of orbital resources while introducing novel concepts, such as the idea of data processing onboard a satellite. Recent market trends suggest novel technologies, such as Deep/Machine Learning and/or Artificial Intelligence, for the space industry. The crucial aspect is hosting and executing diverse software across an abstract hardware platform, which will be re-instantiated regularly. The key component to this objective is virtualization.
In this Diploma Thesis, we investigate two virtualization strategies or we could say two potential candidates, namely the Jailhouse Hypervisor and Docker Containers. Using a Raspberry Pi and Linux we build and configure a fully functional Jailhouse "ecosystem" in order to evaluate the effect of the Jailhouse hypervisor on it and partially compare its performance to that of Docker containers. The study of the Jailhouse Hypervisor on the Raspberry Pi led us to the conclusion that Docker containers add a small amount of system overhead and can be combined with Jailhouse to provide an isolated and secure environment with the flexibility provided by the containerization technique. In terms of observed deterministic behavior, our preliminary findings regarding the effect on real-time performance are quite satisfying. In addition, we confirmed Jailhouse’s strength, isolation, which did not pose a problem in any of our execution scenarios, as well as its weakness, the decrease in application performance when cells communicate or system bus traffic increases. |
en |
heal.advisorName |
Σούντρης, Δημήτριος |
el |
heal.committeeMemberName |
Tsanakas, Panagiotis |
|
heal.committeeMemberName |
Xydis, Sotirios |
|
heal.academicPublisher |
Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών |
el |
heal.academicPublisherID |
ntua |
|
heal.fullTextAvailability |
false |
|