Feedforward Neural Networks for Caching - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles ACM SIGMETRICS Performance Evaluation Review Year : 2019

Feedforward Neural Networks for Caching

Vladyslav Fedchenko
  • Function : Author
  • PersonId : 1061009
Giovanni Neglia
Bruno Ribeiro
  • Function : Author
  • PersonId : 994288

Abstract

We propose a caching policy that uses a feedforward neural network (FNN) to predict content popularity. Our scheme outperforms popular eviction policies like LRU or ARC, but also a new policy relying on the more complex recurrent neural networks. At the same time, replacing the FNN predictor with a naive linear estimator does not degrade caching performance significantly, questioning then the role of neural networks for these applications.
Fichier principal
Vignette du fichier
1810.06930(1).pdf (384.34 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02411461 , version 1 (14-12-2019)

Identifiers

Cite

Vladyslav Fedchenko, Giovanni Neglia, Bruno Ribeiro. Feedforward Neural Networks for Caching: Enough or Too Much?. ACM SIGMETRICS Performance Evaluation Review, 2019, 46 (3), pp.139-142. ⟨10.1145/3308897.3308958⟩. ⟨hal-02411461⟩
24 View
250 Download

Altmetric

Share

Gmail Facebook X LinkedIn More