BLCR: Block-Level Cache Replacement for Large-Scale In-Memory Data Processing Systems

Yingcheng Gu, Yuanhan Du, Huanyu Cheng, Kai Liu, Tuo Cao

2022

Abstract

To reduce the completion time of big data processing applications, in-memory computing has been widely used in today’s systems. Since servers’ memory capacities are typically limited, there is a need to decide which data should be cached in memory, namely the cache replacement problem. However, existing works fall insufficient in analysing the directed acyclic graphs of applications. Moreover, they optimize cache replacement in the resilient distributed data level, which is coarse-grained. In this paper, we investigate the block-level cache replacement problem and formulate it as an integer programming problem. Since it has the optimal substructure property, we develop the algorithm BLCR based on the dynamic programming technique. Trace-driven simulations are conducted to evaluate the performance of BLCR and the results show its superiority over the state-of-the-art alternatives.

Download


Paper Citation


in Harvard Style

Gu Y., Du Y., Cheng H., Liu K. and Cao T. (2022). BLCR: Block-Level Cache Replacement for Large-Scale In-Memory Data Processing Systems. In Proceedings of the 3rd International Symposium on Automation, Information and Computing - Volume 1: ISAIC; ISBN 978-989-758-622-4, SciTePress, pages 256-260. DOI: 10.5220/0011920800003612


in Bibtex Style

@conference{isaic22,
author={Yingcheng Gu and Yuanhan Du and Huanyu Cheng and Kai Liu and Tuo Cao},
title={BLCR: Block-Level Cache Replacement for Large-Scale In-Memory Data Processing Systems},
booktitle={Proceedings of the 3rd International Symposium on Automation, Information and Computing - Volume 1: ISAIC},
year={2022},
pages={256-260},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011920800003612},
isbn={978-989-758-622-4},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Symposium on Automation, Information and Computing - Volume 1: ISAIC
TI - BLCR: Block-Level Cache Replacement for Large-Scale In-Memory Data Processing Systems
SN - 978-989-758-622-4
AU - Gu Y.
AU - Du Y.
AU - Cheng H.
AU - Liu K.
AU - Cao T.
PY - 2022
SP - 256
EP - 260
DO - 10.5220/0011920800003612
PB - SciTePress