Citation
Jorge Ortiz. "A Platform Architecture for Sensor Data Processing and Verification in Buildings". Talk or presentation, 19, November, 2013.

Abstract
Buildings consume 72% of electricity and nearly half of the total energy produced in the United States. In order to improve their performance, deep sampling and analytics must be deployed at scale. However, each building is uniquely designed and highly heterogenous -- limiting portability and preventing widespread deployment of solutions. In this talk, we discuss a platform architecture with two core services: a file system service and a verification service. The former offers a common namespace and standard access method for building sensors/actuators, their data, and data processing jobs. It also allows applications to write portable code. The latter is a verification service that verifies various system behavioral patterns and physical relationships between sensors in order to maintain an accurate representation of the building in software. We evaluate the effectiveness of our architecture and verification techniques through several real-world deployments across 7 buildings in two countries that support over half a dozen applications.

Electronic downloads

• ortiz_DREAMSeminar_PlatformArchSFS_archive.pdf · application/pdf · 14151 kbytes

• HTML

  Jorge Ortiz. <a
  href="http://chess.eecs.berkeley.edu/pubs/1037.html"
  ><i>A Platform Architecture for Sensor Data
  Processing and Verification in
  Buildings</i></a>, Talk or presentation,  19,
  November, 2013.

• Plain text

  Jorge Ortiz. "A Platform Architecture for Sensor Data
  Processing and Verification in Buildings". Talk or
  presentation,  19, November, 2013.

• BibTeX

  @presentation{Ortiz13_PlatformArchitectureForSensorDataProcessingVerification,
      author = {Jorge Ortiz},
      title = {A Platform Architecture for Sensor Data Processing
                and Verification in Buildings},
      day = {19},
      month = {November},
      year = {2013},
      abstract = {Buildings consume 72% of electricity and nearly
                half of the total energy produced in the United
                States. In order to improve their performance,
                deep sampling and analytics must be deployed at
                scale. However, each building is uniquely designed
                and highly heterogenous -- limiting portability
                and preventing widespread deployment of solutions.
                In this talk, we discuss a platform architecture
                with two core services: a file system service and
                a verification service. The former offers a common
                namespace and standard access method for building
                sensors/actuators, their data, and data processing
                jobs. It also allows applications to write
                portable code. The latter is a verification
                service that verifies various system behavioral
                patterns and physical relationships between
                sensors in order to maintain an accurate
                representation of the building in software. We
                evaluate the effectiveness of our architecture and
                verification techniques through several real-world
                deployments across 7 buildings in two countries
                that support over half a dozen applications.},
      URL = {http://chess.eecs.berkeley.edu/pubs/1037.html}
  }
  

Posted by Armin Wasicek on 20 Nov 2013.
Groups: chessworkshop
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.