prikaz prve stranice dokumenta Optimizacija mrežnog prometa u decentraliziranom okruženju
No public access
undergraduate thesis
Optimizacija mrežnog prometa u decentraliziranom okruženju
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. urn:nbn:hr:168:469535
Prpić, Emil
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Prpić, E. (2024). Optimizacija mrežnog prometa u decentraliziranom okruženju (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:469535

Prpić, Emil. "Optimizacija mrežnog prometa u decentraliziranom okruženju." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:469535

Prpić, Emil. "Optimizacija mrežnog prometa u decentraliziranom okruženju." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:469535

Prpić, E. (2024). 'Optimizacija mrežnog prometa u decentraliziranom okruženju', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 02 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:469535

Prpić E. Optimizacija mrežnog prometa u decentraliziranom okruženju [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2024 [cited 2024 December 02] Available at: https://urn.nsk.hr/urn:nbn:hr:168:469535

E. Prpić, "Optimizacija mrežnog prometa u decentraliziranom okruženju", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:168:469535

Metadata
TitleOptimizacija mrežnog prometa u decentraliziranom okruženju
Title (english)Optimization of network traffic in decentralized environment
AuthorEmil Prpić
MentorGordan Ježić (mentor)
Committee memberGordan Ježić (predsjednik povjerenstva)
Committee memberIvan Kralj (član povjerenstva)
Committee memberKatarina Mandarić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2024-07-08, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ovaj rad istražuje optimizaciju mrežnog prometa u decentraliziranim grafovskim neuronskim mrežama (GNN). Decentralizirane GNN-e su ključne za skalabilne i učinkovite distribucije podataka, no s njima dolazi i izazov smanjenja redundancije podataka i mrežnog opterećenja. U ovom istraživanju usporedili smo nekoliko algoritama za kompresiju podataka, uključujući LZMA, Bzip2 i zlib, kako bismo utvrdili najefikasniji pristup za optimizaciju mrežnog prometa u ovom kontekstu. Rezultati eksperimenata pokazali su da zlib algoritam nudi najbolju ravnotežu između brzine kompresije i učinkovitosti smanjenja veličine podataka. Zbog svoje visoke učinkovitosti i niske računalne zahtjevnosti, zlib je izabran za daljnju analizu. Korištenjem DecGNN simulatora, utvrdili smo da primjena zlib algoritma omogućava prosječnu kompresiju od oko 54%, što značajno smanjuje mrežno opterećenje. Ova optimizacija ne samo da smanjuje količinu podataka koji se prenose kroz mrežu, već i poboljšava ukupne performanse mrežnog sustava, smanjujući operativne troškove i povećavajući otpornost na zagušenja i kvarove. Na temelju dobivenih rezultata, preporučuje se daljnje istraživanje i razvoj ovih tehnika kako bi se dodatno unaprijedila njihova primjena u stvarnim uvjetima, te iskoristile sve prednosti koje nude u pogledu skalabilnosti, sigurnosti i robusnosti.
Abstract (english)
This final thesis investigates the optimization of network traffic in decentralized graph neural networks. Decentralized GNNs are crucial for scalable and efficient data distribution, but they also pose the challenge of reducing data redundancy and network load. In this study, we compared several data compression algorithms, including LZMA, Bzip2, and zlib, to determine the most efficient approach for optimizing network traffic in this context. The experimental results showed that the zlib algorithm offers the best balance between compression speed and data size reduction efficiency. Due to its high efficiency and low computational requirements, zlib was selected for further analysis. Using the DecGNN simulator, we found that applying the zlib algorithm achieved an average compression rate of approximately 54%, significantly reducing network load. This optimization not only reduces the amount of data transmitted over the network but also improves the overall performance of the network system, lowering operational costs and increasing resilience to congestion and failures. Based on the obtained results, further research and development of these techniques are recommended to enhance their application in real-world conditions and to leverage all the advantages they offer in terms of scalability, security, and robustness.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:469535
Study programmeTitle: Computing
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2024-10-12 21:55:52