SensorSink Pond is a timeseries database, optimized for sampled sensor data at Big Volume. anticipating the need for scalable sensor solution in the age of Internet of Things that are coming.

• 10s of Millions of sensors per database
• Update speed (sampling rate) from a millisecond to years
• Primary Datatypes
  • Real - a floating point measurement
  • Integer - a 64 bit integer signed number
  • Count - an additive field, typically growing with the value added
  • Boolean - True/False indicator
  • String - a Text field of arbitrary size.
  • JSON - Json objects of arbitrary size.
• Apply functions to timeseries
  • Mininum, Maximum, Average and Standard Deviation over a defined time interval for a timeseries
  • Minimum, Maximum, Average and Standard Deviation on multiple timeseries, single sample within a time window.
  • Custom functions
  • Function applied to, and modifying the timeseries. Or to create new timeseries.
• Query Language for timeseries data
  • Functions
  • Aggregation
  • Massive volumes
  • Time centric
• REST interface
  • JSON as primary representation
  • XML and BSON supported from SensorSink
  • Custom representations must be possible.
  • GZIP support
  • Multi-query aggregation into GZIP result file.