Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
Journal of the ACM (JACM)
The Real-Time Specification for Java
The Real-Time Specification for Java
FUNDAMENTAL DESIGN PROBLEMS OF DISTRIBUTED SYSTEMS FOR THE HARD-REAL-TIME ENVIRONMENT
FUNDAMENTAL DESIGN PROBLEMS OF DISTRIBUTED SYSTEMS FOR THE HARD-REAL-TIME ENVIRONMENT
LITMUS^RT: A Testbed for Empirically Comparing Real-Time Multiprocessor Schedulers
RTSS '06 Proceedings of the 27th IEEE International Real-Time Systems Symposium
A real-time Java virtual machine with applications in avionics
ACM Transactions on Embedded Computing Systems (TECS)
Multiprocessors and the Real-Time Specification for Java
ISORC '08 Proceedings of the 2008 11th IEEE Symposium on Object Oriented Real-Time Distributed Computing
JEOPARD: Java environment for parallel real-time development
JTRES '08 Proceedings of the 6th international workshop on Java technologies for real-time and embedded systems
Proceedings of the 7th International Workshop on Java Technologies for Real-Time and Embedded Systems
A framework for flexible scheduling in the RTSJ
ACM Transactions on Embedded Computing Systems (TECS)
| Hi-index | 0.00 |
In this paper, we present a framework for various multiprocessor scheduling algorithms by minimal modification of current Real-Time Specification for Java (RTSJ) [6]. Although the current version of RTSJ provides a secure platform and rich functionalities for real-time Java applications, it lacks multiprocessor support mechanisms, e.g., absence of functions to support processor affinity, to efficiently utilize multiple processing resources. For this reason, we establish a multiprocessor-aware scheduling framework by using system calls of operating systems to make use of processor affinity, FIFO scheduler, scheduling parameter settings, and precision sleep timer functions. In addition to the framework, we also take categorization taxonomy introduced by Carpenter et al. in [1], which generalizes multiprocessor scheduling algorithms on two criteria of migration degrees and priority change complexity. Then our experimental evaluation on the framework with each scheduler class in the categorization taxonomy shows the framework's runtime overhead, which proves the feasibility of our implementation.