Middleware 2009

Volume 5896 of the series Lecture Notes in Computer Science pp 328-348

Persistent Temporal Streams

  • David HilleyAffiliated withCollege of Computing, Georgia Institute of Technology
  • , Umakishore RamachandranAffiliated withCollege of Computing, Georgia Institute of Technology


Distributed continuous live stream analysis applications are increasingly common. Video-based surveillance, emergency response, disaster recovery, and critical infrastructure protection are all examples of such applications. They are characterized by a variety of high- and low-bandwidth streams as well as a need for analyzing both live and archived streams. We present a system called Persistent Temporal Streams (PTS) that supports a higher-level, domain-targeted programming abstraction for such applications. PTS provides a simple but expressive stream abstraction encompassing transport, manipulation and storage of streaming data. In this paper, we present a system architecture for implementing PTS. We provide an experimental evaluation which shows the system-level primitives can be implemented in a lightweight and high-performance manner, and an application-based evaluation designed to show that a representative high-bandwidth stream analysis application can be implemented relatively simply and with good performance.