PRESTO - A Predictive Storage Architecture for Sensor Networks

Presto - suddenly as if by magic

Overview

Recent years have seen a tremendous growth in extending the reaches of the Internet to numerous sensor data sources including RFIDs, weather and habitat monitors, building monitors, remote sensing data such as radar and others. These sensor data sources span a spectrum of power, data rate and platform requirements, from passive RFIDs and battery-powered wireless sensors to high bandwidth radar nodes.

A unifying constraint across the spectrum of sensor data management applications is that the remote sensor nodes at the network edge are usually constrained in power, functionality and/or bandwidth, and communicate using multi-hop routing to a resource-rich proxy that connects the sensors to the Internet. PRESTO takes a fresh look at the design of tiered large-scale sensor networks that comprise tethered and untethered elements and addresses three key questions:

Where should sensor data be archived?
How can low-latency, interactive query processing be supported in-spite of the power and bandwidth constraints of sensors, frequent sensor failures and vagaries of wireless links?
How can a user access data generated at numerous sensors, and across a large geographic domain in an efficient manner?

There are five main characteristics of PRESTO

Tiered Design: PRESTO reflects our philosophy that scalable sensor networks of the future will have multiple tiers, with a several tens of untethered sensors per tethered sensor proxy and several tens of sensor proxies per application.
Predictive Storage: PRESTO makes extensive use of predictive techniques that are a natural fit to the correlated behavior of the physical world. We exploit technology trends in storage to build an architecture that emphasizes archival at remote sensors and extensive modeling and predictive caching at proxies.
Interactive Use: PRESTO is designed for rapid responses for users posing ad-hoc queries on distributed sensor data without incurring the energy costs of the data streaming approach or the losses, latency and reliability concerns of directly querying remote sensors.
Archival Queries: PRESTO supports archival queries on data that may be deemed to be interesting post-facto. The ability to query historical data is important in many sensor applications to conduct postmortems of unexpected and unusual events to better understand them for the future.
Single Logical View of Data: PRESTO aims to provide a single logical view of data distributed across many sensor proxies and numerous remote sensors. Such a view can abstract the user from the variabilities at many levels---lossy and unreliable remote sensor network; spatial and temporal consistency issues in the sensor data; as well as bandwidth and connectivity issues in the case of wireless proxies.

Papers

Ming Li, Deepak Ganesan and Prashant Shenoy, PRESTO: Feedback-Driven Data Management in Sensor Networks, Proceedings of the Third ACM/USENIX Symposium on Networked Systems Design and Implementation (NSDI), San Jose, CA, May 2006.
Gaurav Mathur, Peter Desnoyers, Deepak Ganesan and Prashant Shenoy, Ultra-Low Power Storage for Sensor Networks, Proceedings of IEEE/ACM Conference on Information Processing in Sensor Networks - Track on Sensor Platform, Tools and Design Methods for Networked Embedded Systems (IPSN-SPOTS), Nashville, TN, April 2006.
Peter Desnoyers, Deepak Ganesan and Prashant Shenoy, TSAR: A Two Tier Storage Architecture Using Interval Skip Graphs, Proceedings of the Third ACM Conference on Embedded Networked Sensor Systems (SenSys), San Diego, CA. November 2005
Peter Desnoyers, Deepak Ganesan, Huan Li and Prashant Shenoy, PRESTO: A Predictive Storage Architecture for Sensor Networks, Proceedings of the Tenth Workshop on Hot Topics in Operating Systems (HotOS X), Santa Fe, New Mexico, June 2005.

People

Funding

This work was supported by a grant from the Engineering Research Centers program of the National Science Foundation under cooperative agreement EEC-0313747 and NSF CAREER Award CNS-0546177.