Revisiting the "perc real-time API"

  • Authors:
  • Kelvin Nilsen

  • Affiliations:
  • Atego Systems, San Diego, CA

  • Venue:
  • Proceedings of the 10th International Workshop on Java Technologies for Real-time and Embedded Systems
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Long before formation of the Java Community Process, a paper discussing issues of real-time Java and a companion paper proposing a draft API for real-time Java were distributed via a web server hosted by Iowa State University [1, 2]. Nearly nine hundred copies of the draft API were downloaded in the first eight months following its release in January 1996. Industry response was overwhelmingly enthusiastic, with multiple real-time operating system companies approaching the author of the papers with offers to fund implementation of the proposed concepts. Ultimately, the author left Iowa State University to found NewMonics, with initial financial backing from an angel investor who helped the company eventually receive more than $15 million of venture capital funding. But the fortunes of real-time Java have not lived up to the promise of the initial enthusiasm. Now, over fifteen years after the initial real-time Java reports were published, and over ten years after initial publication of the Real-Time Specification for Java (RTSJ) [3], real-time Java is still an oddity within the domains of embedded real-time computing. There are only a relatively small number of reported real-time Java success stories. And the only two projects based on the RTSJ have been defense applications, with severe limitations on information about their experience. In the past few years, both IBM and Oracle have backed away from their originally aggressive support for the RTSJ. In recent months, Atego has been approached by an aerospace customer who is requesting access to the technologies described in the original Perc Real-Time API (PRTAPI) document that had been first published in 1996. A review of that original API document serves to remind the real-time Java community of the ideals that initially attracted such strong industry interest in the promise of real-time Java. This paper describes a possible modern incarnation of the original PRTAPI and discusses why its features are considered preferable to currently available real-time Java technologies.