Post-graduate theses
Current Record: 95 of 824
|
Identifier |
000438390 |
Title |
LuaGuardia: a confidential computing framework for trusted execution environments |
Alternative Title |
LuaGuardia: ένα σύστημα εκτέλεσης κρίσιμου κώδικα για περιβάλλοντα ασφαλούς εκτέλεσης |
Author
|
Καρνίκης, Δημήτριος Ε.
|
Thesis advisor
|
Πρατικάκης, Πολύβιος
|
Reviewer
|
Ιωαννίδης, Σωτήρης
Δημητρόπουλος, Ξενοφώντας
|
Abstract |
Confidential computing applications are enabled by Trusted Execution environments (TEEs)
that are becoming increasingly widespread in the computing landscape. However, their
development and deployment remains challenging due to several reasons. The lack of
high-level TEE abstractions complicates application development and forces the use of
low-level memory- and type-unsafe abstractions. These challenges are exacerbated by
technical issues regarding runtime extensibility, management of cryptographic operations,
and restricted interfaces: even porting existing applications requires manual partitioning,
re-compilation, and linking steps.
This work presents LuaGuardia, a system simplifying the development of confidential computing. LuaGuardia addresses the aforementioned challenges by offering a set of
abstractions around a TEE-embedded runtime environment of a high-level programming
language. LuaGuardia’s abstractions simplify the development and deployment of such
applications in a type- and memory-safe manner. It also offers a runtime library solving
technical challenges such as code signing, system-call offloading, access control, and dynamic code loading. A series of optimizations is also provided that accelerate protected
code execution. Our evaluation applies LuaGuardia to a diverse set of applications, cryptographic functions but also real-world commercial applications, with an average overhead of 18%, the majority of which is due to I/O delays.
|
Language |
English |
Subject |
Code offloading |
|
Intel SGX |
|
Secure computing |
|
Security |
|
TEE |
|
Ασφαλείς υπολογισμοί |
|
Δυναμική εκτέλεση κώδικα |
|
ΠΑΕ |
Issue date |
2021-03-26 |
Collection
|
School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses
|
|
Type of Work--Post-graduate theses
|
Permanent Link |
https://elocus.lib.uoc.gr//dlib/4/e/b/metadata-dlib-1615984474-678758-20001.tkl
|
Views |
520 |