PERRY: Flexible and Scalable Data Preprocessing System for ”ML for Networks” Pipelines
Published in UC Santa Barbara Proquest Electronic Theses and Dissertation, 2023
PERRY is my masters thesis dissertation in which I discuss the prevailing issue of tight coupling between data processing and model training parts in Machine learning for networking pipelines and then try to address this problem with a flexible and scalable data processing framework called PERRY. Leveraging state of the art tools and being scalable with limited resources makes our framework an easy to use solution for networking researchers.
Recommended citation: Battula, N. (2023). PERRY: A Flexible and Scalable Data Preprocessing System for "ML for Networks" Pipelines. UC Santa Barbara. ProQuest ID: Battula_ucsb_0035N_16163. Merritt ID: ark:/13030/m5j49rxb. Retrieved from https://escholarship.org/uc/item/6p42s4w9
Download Paper | Download Slides