Sensor Network Research

1
Sensor Network Research
University of Massachusetts, Amherst
Ultra-Low Power Data Storage for Sensors Gaurav
Mathur, Peter Desnoyers, Deepak Ganesan and
Prashant Shenoy
Introduction Each generation of sensor
platforms reduces computation and communication
costs, but storage costs have not tracked this
trend.

• Applications working with high data volumes
• Camera sensor networks
• Acoustic
• Tracking networks for animals, vehicles, etc.
• Seismic
• Biological sensor networks

What is the most energy-efficient storage
platform for sensor devices ? How does the
energy cost of storage compare to that of
computation and communication ? What are the
implications of an ultra-low power storage
subsystem on sensor net design ?
Measurement Study of Storage Technology
NAND flash based Storage Framework for Sensor
Devices
Toshiba Parallel NAND flash adapter

• Motivation
• Existing storage interfaces
• File systems - YAFFS, ELF, YAFFS2 But sensors
  do not operate on files
• Stream storage and indexing MicroHash
• Other sensor application requirements
• Easy access to archived data for query processing
• Support for data consistency
• Hardware independence
• Overcome memory constraints of platforms

Mica2 Atmel serial NOR flash
TelosB STM serial NOR flash
MMC adapter

• Compare MMC, NOR NAND flash technology
• Measure active sleep mode power consumption

Comparison of the per byte cost of operations
Without ECC. Cost of performing ECC in software
is approx 0.026uJ/byte
Comparison of computation, communication and
storage costs

• Features
• Storage object interface exposed to applications
  stream, stack, queue, index, array
• Implemented on NAND flash
• Support concurrent data streams
• Minimal memory footprint
• Transaction support (atomic and durable)
• Data prioritization

• Parallel NAND flash most energy efficient
• More than 200 times less in comparison to
  communication costs
• Comparable to computation costs
• Enables ultra-low power, almost infinite
  storage (1 GB) for sensors

http//sensors.cs.umass.edu/projects/essense