Functional decomposition and disaggregation of XR applications into programmable devices

• Grant number: 23/06124-5
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: August 01, 2023
• End date: July 31, 2027
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Fabio Luciano Verdi
• Grantee: Guilherme Mendes Vieira de Matos
• Host Institution: Centro de Ciências em Gestão e Tecnologia (CCGT). Universidade Federal de São Carlos (UFSCAR). Campus de Sorocaba. Sorocaba , SP, Brazil
• Company: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação (FEEC)
• Associated research grant: 21/00199-8 - SMART NEtworks and ServiceS for 2030 (SMARTNESS), AP.PCPE

Abstract

The evolution of software and hardware reached the point where CPUs can no longer meet the new demands of the future. To deal with this problem, new devices and paradigms have emerged, and among these devices are various programmable network devices, like Tofino switches, SmartNICs and FPGAs, which are used to perform network processing and other specialized functions, thus releasing the CPU to only handle applications. Still in an attempt to deal with the initial problem, disaggregation was proposed, presenting the concept where resources are separated into pools, facilitating elasticity, deployment and management. Naturally, the disaggregation concept covers the programmable network devices, and currently the great challenge in the union between these two concepts is to carry out the functional decomposition of applications into blocks of codes and the mapping of these blocks to different hardwares. However, current work mainly focuses on offloading network functions to programmable network devices, especially SmartNICs. This PhD aims to advance the field by applying the concepts of functional decomposition and mapping both network functions and extended reality (XR) applications to programmable hardwares. In this effort, this work proposes to (1) design a solution that allows functional decomposition and (2) map these functional blocks to programmable hardware based on the specific characteristics of each function and also of the hardware, thus achieving the best match between function and device. It is expected that the PhD project will generate scientific contributions in the areas of resource disaggregation and hardware offloading. (AU)