Fuzzy-token: An adaptive Mac protocol for wireless network-on-chip

Recent computer architecture trends herald the arrival of massive multiprocessors with more than a thousand processor cores within a single chip. In this context, as parallel programs continue to increase the amount of data sharing and signaling between cores, on-chip communication becomes a critical issue. Unfortunately, traditional on-chip networks have been proven to not scale well in terms of latency or energy consumption, slowing down the computation in thousand-core processors. The Wireless Network-on-Chip (WNoC) paradigm holds considerable promise for the implementation of on-chip networks that will enable such massive multicore chips. One of the main challenges, however, resides in the design of methods that provide fast and efficient access to the wireless channel while adapting to the constant traffic changes within and across applications. Existing approaches are either cumbersome or do not provide the required adaptivity. We propose Fuzzy-Token, a simple protocol that leverages the unique properties of the on-chip scenario to deliver efficient and low-latency access irrespective of the application characteristics. We substantiate our claim via simulations with a synthetic traffic suite and real application traces.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Franques Garcia, Antonio Maria

Contributors dc:contributor

• Torrellas, Josep

• Chip Multiprocessor
• Wireless Network-on-Chip
• Media Access Control

Statement dc:rights

• Copyright 2019 Antonio Maria Franques Garcia

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106374

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/106374