Loading...

WiP: Floating xy-yx: An efficient thermal management routing algorithm for 3d nocs

Safari, M ; Sharif University of Technology | 2018

596 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/DASC/PiCom/DataCom/CyberSciTec.2018.00128
  3. Publisher: Institute of Electrical and Electronics Engineers Inc , 2018
  4. Abstract:
  5. 3D Network-on-Chips (3D NoCs) have higher scalability, higher throughput, and lower power consumption over 2D NoCs. However, the reliability of data transfer in 3D NoCs is seriously threatened by thermal problems. This is due to poor heat dissipation, inappropriate traffic distribution, and cooling restriction for layers away of the chip heat-sink in 3D NoCs. To solve this problem, this paper proposes an efficient deadlock-free and traffic-And thermal-Aware routing algorithm, called Floating XY-YX. The main idea behind Floating XY-YX routing algorithm is twofold: 1) to use XY and YX routing algorithms in consecutive layers in dessicate form, and 2) to evenly load the traffic, which is realized by switching the type of XY and YX routing algorithms to YX and XY routing algorithms, respectively. Based on the experimental results, the proposed routing algorithm reduces the temperature variance by 30.74% and traffic variance by 23.77% with respect to the state-of-The-Art routing mechanisms. © 2018 IEEE
  6. Keywords:
  7. 3D NoC ; Routing Algorithm ; Thermal Problems ; Big data ; Binary alloys ; Data transfer ; Heat problems ; Problem solving ; Routing algorithms ; Thermal management (electronics) ; Thermoanalysis ; Deadlock free ; Lower-power consumption ; Routing mechanism ; State of the art ; Temperature variance ; Thermal-Aware ; Traffic distributions ; XY routing algorithms ; Network-on-chip
  8. Source: 16th IEEE International Conference on Dependable, Autonomic and Secure Computing, IEEE 16th International Conference on Pervasive Intelligence and Computing, IEEE 4th International Conference on Big Data Intelligence and Computing and IEEE 3rd Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2018, 12 August 2018 through 15 August 2018 ; 2018 , Pages 730-735 ; 9781538675182 (ISBN)
  9. URL: https://ieeexplore.ieee.org/document/8511971