logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-09222021-115659


Tipo di tesi
Tesi di laurea magistrale
Autore
SAHBAEI KEYA, SAHAR
Indirizzo email
s.sahbaeikeya@studenti.unipi.it, sahar.sahbaee@gmail.com
URN
etd-09222021-115659
Titolo
survey of state-of-the-art programming frameworks for graphics processing units, GPUs
Dipartimento
INFORMATICA
Corso di studi
INFORMATICA
Relatori
relatore Prof. Danelutto, Marco
Parole chiave
  • programming frameworks for GPUs
Data inizio appello
08/10/2021
Consultabilità
Tesi non consultabile
Riassunto
To begin, a brief introduction of the graphics processing unit (GPU) and parallel computing. Afterwards, we look into the three most popular programming platforms for GPUs in greater depth, and compare them on various criteria such as architectural model, programming model and so on, by demonstrating their similarity, differences, advantages and disadvantages in their code. Finally, we explain three of the most common parallel computing patterns, and then compare the three platforms by programming the patterns with using each of the three application frameworks. Finally, we have reached a decision.
File