Learning-based routing algorithms for on-chip networks

3Citations
Citations of this article
4Readers
Mendeley users who have this article in their library.
Get full text

Abstract

In this chapter, we investigate highly adaptive routing algorithms for balancing the traffic over the network based on learning approaches. The proposed methods aim to provide up-to-dated local and global congestion information at each switch. At first, the learning method is applied to a network utilizing the minimal routing. In low traffic loads, minimal methods can achieve optimal performance, while they are inefficient in avoiding hotspots when the traffic load increases. The reason of this inefficiency is that minimal methods can propagate messages through at most two directions at each switch. When the shortest paths are congested, sending more messages through them can deteriorate the congestion condition considerably. In order to address this issue, we present a non-minimal routing algorithm for on-chip networks that provides a wide range of alternative paths between each pair of source and destination switches. Initially, the algorithm determines all permitted turns in the network including 180-degree turns on a single channel without creating cycles. The implementation of the algorithm provides the best usage of all allowable turns to route messages more adaptively in the network. On top of that, for selecting a less congested path, an optimized and scalable learning method is utilized. The learning method is based on local and global congestion information and can estimate the latency from each output channel to the destination region.

Cite

CITATION STYLE

APA

Ebrahimi, M., & Daneshtalab, M. (2014). Learning-based routing algorithms for on-chip networks. In Routing Algorithms in Networks-on-Chip (Vol. 9781461482741, pp. 105–125). Springer New York. https://doi.org/10.1007/978-1-4614-8274-1_5

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free