FAST TRANSPARENT VIRTUAL MEMORY FOR COMPLEX DATA PROCESSING IN SENSOR NETWORKS

Nan Lin, Yabo Dong, Dongming Lu

2012

Abstract

Sensor networks has been used in numerous fields, some of which require complex processing of large amount of data, such as visual surveillance. Low-power MCUs on sensor nodes have kilobytes of RAM which is not enough for mass data processing. In this paper, we present a transparent software-based virtual memory for mass complex data processing in sensor networks which can be much larger than the physical memory. Traditional optimizations have been employed and tuned to best suit the virtual memory. Address translations were found to be a major overhead in virtual memory for most applications. With respect to assembly conversion commonly used in other software-based virtual memory systems, we use flexible C code transformation which generates cache index buffers to greatly reduce the address translation overhead. The usability of the virtual memory is further verified by a number of common algorithms using the virtual memory.

Paper Citation

in Harvard Style

Lin N., Dong Y. and Lu D. (2012). FAST TRANSPARENT VIRTUAL MEMORY FOR COMPLEX DATA PROCESSING IN SENSOR NETWORKS . In Proceedings of the 1st International Conference on Sensor Networks - Volume 1: SENSORNETS, ISBN 978-989-8565-01-3, pages 24-34. DOI: 10.5220/0003804900240034

in Bibtex Style

@conference{sensornets12,
author={Nan Lin and Yabo Dong and Dongming Lu},
title={FAST TRANSPARENT VIRTUAL MEMORY FOR COMPLEX DATA PROCESSING IN SENSOR NETWORKS},
booktitle={Proceedings of the 1st International Conference on Sensor Networks - Volume 1: SENSORNETS,},
year={2012},
pages={24-34},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003804900240034},
isbn={978-989-8565-01-3},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 1st International Conference on Sensor Networks - Volume 1: SENSORNETS,
TI - FAST TRANSPARENT VIRTUAL MEMORY FOR COMPLEX DATA PROCESSING IN SENSOR NETWORKS
SN - 978-989-8565-01-3
AU - Lin N.
AU - Dong Y.
AU - Lu D.
PY - 2012
SP - 24
EP - 34
DO - 10.5220/0003804900240034