G
Georgios Zacharopoulos
Researcher at University of Lugano
Publications - 10
Citations - 76
Georgios Zacharopoulos is an academic researcher from University of Lugano. The author has contributed to research in topics: Computer science & Compiler. The author has an hindex of 4, co-authored 6 publications receiving 41 citations.
Papers
More filters
Proceedings ArticleDOI
Machine Learning Approach for Loop Unrolling Factor Prediction in High Level Synthesis
TL;DR: This work proposes a novel Machine Learning methodology, able to jointly forecast the optimal loop unrolling factors for all the loops in a target application, and demonstrates that this method results in a better prediction score and a reduced convergence time compared to other state-of-the-art approaches.
Proceedings ArticleDOI
Multiversioned decoupled access-execute: the key to energy-efficient compilation of general-purpose programs
Konstantinos Koukos,Per Ekemark,Georgios Zacharopoulos,Vasileios Spiliopoulos,Stefanos Kaxiras,Alexandra Jimborean +5 more
TL;DR: This work proposes a universal compile-time method to decouple general-purpose applications, using simple but efficient heuristics and shows that automatic decoupled execution significantly reduces the energy expenditure of irregular or memory-bound applications and even yields slight performance boosts.
Proceedings ArticleDOI
Compiler-Assisted Selection of Hardware Acceleration Candidates from Application Source Code
Georgios Zacharopoulos,Lorenzo Ferretti,Giovanni Ansaloni,Giuseppe Di Guglielmo,Luca P. Carloni,Laura Pozzi +5 more
TL;DR: This work presents a framework for automatically identifying software segments that are promising candidates for hardware acceleration and to evaluate, from un-modified software code, the potential speedup and resource requirements, based on Intermediate Representation (IR) analysis passes, which is embedded in the LLVM compiler toolchain.
Journal ArticleDOI
RegionSeeker: Automatically Identifying and Selecting Accelerators From Application Source Code
TL;DR: A method to identify subgraphs of control flow graphs having a single input control point and a single output control point that are good targets for the synthesis of application specific hardware accelerators and an LLVM-based toolchain that, analyzing a software application, automatically selects its most profitable regions given an area constraint is provided.
Employing Hardware Transactional Memory in Prefetching for Energy Efficiency
TL;DR: Energy efficiency is becoming a highly significant topic regarding modern hardware and the need for decreased energy consumption in computers and more battery life in laptops and smart-phones is becoming more significant.