MUSTACHE: Multi-Step-Ahead Predictions for Cache Eviction

11/03/2022
by   Gabriele Tolomei, et al.
0

In this work, we propose MUSTACHE, a new page cache replacement algorithm whose logic is learned from observed memory access requests rather than fixed like existing policies. We formulate the page request prediction problem as a categorical time series forecasting task. Then, our method queries the learned page request forecaster to obtain the next k predicted page memory references to better approximate the optimal Bélády's replacement algorithm. We implement several forecasting techniques using advanced deep learning architectures and integrate the best-performing one into an existing open-source cache simulator. Experiments run on benchmark datasets show that MUSTACHE outperforms the best page replacement heuristic (i.e., exact LRU), improving the cache hit ratio by 1.9 required to handle cache misses by 18.4

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/05/2010

FIFO anomaly is unbounded

Virtual memory of computers is usually implemented by demand paging. For...
research
12/20/2019

CacheQuery: Learning Replacement Policies from Hardware Caches

We show how to infer deterministic cache replacement policies using off-...
research
03/23/2018

Unix Memory Allocations are Not Poisson

In multitasking operating systems, requests for free memory are traditio...
research
10/06/2022

Paging with Succinct Predictions

Paging is a prototypical problem in the area of online algorithms. It ha...
research
05/21/2017

MITHRIL: Mining Sporadic Associations for Cache Prefetching

The growing pressure on cloud application scalability has accentuated st...
research
05/29/2022

TransforMAP: Transformer for Memory Access Prediction

Data Prefetching is a technique that can hide memory latency by fetching...
research
08/15/2015

Cracking Intel Sandy Bridge's Cache Hash Function

On Intel Sandy Bridge processor, last level cache (LLC) is divided into ...

Please sign up or login with your details

Forgot password? Click here to reset