Javascript must be enabled for the correct page display

Stateful data analytics over programming models of networks

Pojoga, Gheorghe (2021) Stateful data analytics over programming models of networks. Bachelor's Thesis, Computing Science.

[img]
Preview
Text
bCS_2021_PojogaG.pdf

Download (680kB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (121kB)

Abstract

The optimization of network data processing is a scalable approach for meeting the continuously growing Quality of Service requirements. A solution in this regard is offloading some end-host operators into the network devices, also known as in-network computing. This approach is supported by multiple emerging technologies, that aim at providing more flexibility in terms of network programmability, while preserving the line rate. Although, the stateless packet processing has already found multiple applications across the industry, the implementation of stateful operators has proven to be more challenging due to the limitations of the available hardware. In this paper we will focus on building an understanding on the viability and acceleration capabilities of executing stateful data analytics operators inside the network, by analyzing several concrete problems and their solutions across some of the available network processing frameworks.

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Koldehofe, B. and Boughzala, B.
Degree programme: Computing Science
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Sep 2021 11:45
Last Modified: 15 Sep 2021 11:45
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/26087

Actions (login required)

View Item View Item