Abstract
In this paper, we present DEEPCACHE a novel Framework for content caching, which can significantly boost cache performance. Our Framework is based on powerful deep recurrent neural network models. It comprises of two main components: i) Object Characteristics Predictor, which builds upon deep LSTM Encoder-Decoder model to predict the future characteristics of an object (such as object popularity) - to the best of our knowledge, we are the first to propose LSTM Encoder-Decoder model for content caching; ii) a caching policy component, which accounts for predicted information of objects to make smart caching decisions. In our thorough experiments, we show that applying DEEPCACHE Framework to existing cache policies, such as LRU and k-LRU, signiicantly boosts the number of cache hits.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 64-69 |
| Number of pages | 6 |
| Journal | Computer Communication Review |
| Volume | 48 |
| Issue number | 5 |
| DOIs | |
| State | Published - Oct 2018 |
Bibliographical note
Funding Information:This chesearr was supported in part by US NSF grant CNS-1411636, CNS-1618339 and CNS-1617729, TRAD grant TRA1-14-1-0040,HD and a eiHuaw gift.
Funding Information:
This research was supported in part by US NSF grant CNS-1411636, CNS-1618339 and CNS-1617729, DTRA grant HDTRA1-14-1-0040, and a Huawei gift.
Publisher Copyright:
© 2018 E-flow ACM (Association for Computing Machinery). All Rights Reserved.
Keywords
- Cache hit
- Caching
- Deep learning
- DeepCache
- Fake requests
- Lstm
- Machine learning
- Popularity prediction
- Prefetching
- Proactive caching
- Seq2seq
- Smart caching policies
- Video object caches