ViSE: Virtualized Sensing Environment

1
ViSE Virtualized Sensing Environment

• David Irwin, Mike Zink, Prashant Shenoy
• Jim Kurose, and Deepak Ganesan
• GENI Spiral 1 Cluster D/Orca

2
Background

• GENI and Sensor Networks
• in 10 years, most computerswill be small
  sensors and actuatorsto monitor health, traffic,
  weather, pollution, science experiments,
  surveillance, etc.
• It would seem odd if in 10 years we were still
  living with an Internet that did not take into
  account the needs of the majority of the
  computers then deployed.

3
Sensing and Actuators

• ac-tu-a-tor n. A mechanism that causes a device
  to be turned on or off, adjusted or moved.
• Nearly every device has actuator(s)
• E.g., CPUs, NICs, disks, sensors
• Essence of deeply programmable
• Actuation key for sensors
• Determines data type, quality, quantity, etc.
• E.g., Sampling rate, Steering, Power
• Indirectly affects resource usage
• E.g., energy, bandwidth, storage, processing

4
Broad Range of Sensor Networks
habitat monitoring
microclimate monitoring
CASA-NSF ERC
vehicle tracking in sensor field
animal tracking
radar/weather
video surveillance
auto traffic monitoring
satellite observation (EODIS)
underwater sensing
network traffic monitoring
Many differences, but also many commonalities
5
ViSE Testbed Overview

• Experiment with Actuation
• In addition to slivering CPU, memory, network,
  etc.
• .goal is to virtualize and sliver sensors
• Infrastructure
• 3 nodes - weatherproof, solar-powered
• Network - long-distance 802.11b/g
• Sensors - radar, Weather Station, Camera
• Backplane - cellular embedded node

6
Example Weather Radars

• Sparse, high-power radar
• sensing gap earth curvature effects prevent 72
  of the troposphere below 1 km from being observed
• coarse resolution

10,000 ft
3.05 km
snow
wind
3.05 km
tornado
earth surface
Horz. Scale 1 50 km Vert. Scale 1 -- 2 km
0
40
80
120
160
200
240
RANGE (km)

There is insufficient knowledge about what is
actually happening (or is likely to happen) at
the Earths surface where people live. NRC 1998
7
Benefits of Better Actuation
instead of this.
This

• finer spatial resolution
• beam focus more energy into sensed volume
• multiple looks sense volume with most
  appropriate radars

8
Real-world CASA Example
4/10/07 first CASA data citation by NWS
FLUS74 KOUN 102343 AWUOUN AREA WEATHER UPDATE
NATIONAL WEATHER SERVICE NORMAN OK 742 PM CDT
TUE APR 10 2007 ..WARNING DECISION UPDATE
THIS WARNING DECISION UPDATE CONCERNS COMANCHE
AND GRADY COUNTIES. MESOCYCLONE NEAR STERLING
CONTINUES TO STRENGTHEN PER TWO RADAR VIEWS. CASA
NETWORK ALSO SHOWING PRONOUNCED HOOK. STORM WILL
ENCOUNTER WARM FRONT...WITH POSSIBLE ENHANCED LOW
LEVEL SHEAR JUST EAST OF STERLING AND WEST OF
RUSH SPRINGS. TORNADO WARNING IS POSSIBLE IF NOT
LIKELY TO BE ISSUED AS STORM REACHES SOUTHWEST
GRADY COUNTY. BURKE
Note not policy
9
Orca Cluster D

• Orca Provides
• Basic GENI Structure/Interfaces
• Federation
• Basic VM and Slivering Mechanisms (Xen)
• Basic Scheduling Policies
• Management/Portal Infrastructure

10
Orca Cluster D

• Vise Focus
• Physical Infrastructure
• Virtualized Actuators (Xen)
• Orca Integration
• New VM/Slivering Mechanisms
• Augmented Policies
• Publicly-available Testbed

11
Integration Plans

• ViSE is a self-contained testbed
• Federate through Orca clearinghouse
• Some interaction with other Cluster D projects
• New slivering mechanisms/policies
• Integrate into Orca framework
• Experiment/portal services
• Slice controllers, Gush (Williams)
• Interfaces
• Experiment/Data Plane virtual sensor interface
• Control Plane GENI/Orca
• Operations Management GENI/Orca

12
Timeline Year 1

• Demo - GEC4 (March 09)
• Infrastructure deployed and operational
• Reflectivity Overlay using Google Maps
• Initial Orca Integration
• Web portal request slices of Xen VMs
• Demo - GEC5 (July 09)
• End-to-end slices across ViSE and compute cluster

13
Questions?