| Katcher, Daniel I., Jay K. Strosnider, and Elizabeth A. Hinzelman-Fortino. "Dynamic versus Fixed Priority Scheduling: A Case Study" http://usa.ece.cmu.edu/Jteam/papers/abstracts/tse93.abs.html . |
....load balancing. Programs must be able to recognize that current resources are no longer available. Schedulers are employed in the presence of faults to manage resources against program needs using dynamic or fixed priority scheduling for timing correctness of critical application tasks. 1] [2], 3] and [4] Information and Data Fusion Applications Information and data fusion applications have been a general thrust of research in our lab. The need for faster, cheaper, more efficient processing for fusion has motivated much of our research. Information fusion is the analysis and ....
Katcher, Daniel I., Jay K. Strosnider, and Elizabeth A. Hinzelman-Fortino. "Dynamic versus Fixed Priority Scheduling: A Case Study" http://usa.ece.cmu.edu/Jteam/papers/abstracts/tse93.abs.html .
....task sets respond to scaling that scaling has a bigger effect on more balanced task sets than on more skewed task sets where the behavior of one class may dominate that of all the other classes. We comment briefly on the results we obtained by studying a multimedia task set described in [5]. We changed some of the periods of the original task set to obtain a regeneration cycle of manageable duration. This modified task set had a nominal utilization of 41.3 . At this utilization, the average difference in the numbers of preemptions between ED and RM was negligible while the maximum ....
D.I. Katcher and J.K. Strosnider, "Dynamic versus Fixed Priority Scheduling: A Case Study". Submitted to IEEE Trans. on Software Engineering. 10
....1.48 13.53 3.00 1.1693 74.71 1.88 13.41 3.69 1.0230 78.86 2.42 13.58 4.99 0.9240 83.01 3.04 21.12 6.37 0.8303 87.16 3.82 21.40 7.77 0.7290 91.31 4.90 22.13 9.73 0.6365 92.14 5.18 21.60 10.27 0. 6190 We now comment briefly on the results we obtained by studying a multimedia task set described in [37]. We changed some of the periods of the original task set [37] to obtain a regeneration cycle of manageable duration. This modified task set shown in Table 3.5 where all the periods are in CPU cycles normalized with respect to the requirements of a 9.6 KHz analog interface controller [37] The ....
....2.42 13.58 4.99 0.9240 83.01 3.04 21.12 6.37 0.8303 87.16 3.82 21.40 7.77 0.7290 91.31 4.90 22.13 9.73 0.6365 92.14 5.18 21.60 10.27 0. 6190 We now comment briefly on the results we obtained by studying a multimedia task set described in [37] We changed some of the periods of the original task set [37] to obtain a regeneration cycle of manageable duration. This modified task set shown in Table 3.5 where all the periods are in CPU cycles normalized with respect to the requirements of a 9.6 KHz analog interface controller [37] The task set has a nominal utilization of 41.3 . At this utilization, ....
[Article contains additional citation context not shown here]
Katcher, D., and Strosnider, J., "Dynamic versus Fixed Priority Scheduling: A Case Study". Submitted to IEEE Trans. on Software Engineering.
....a primary memory context switch. Since we do not know the exact thread invocation pattern, we must 10 Note that a practical implementation of the thread dispatcher introduces much less overhead for a static priority policy than for an EDF policy. As have been discussed in several research papers [34, 9], the net result of these two effects is that both schemes deliver comparable performance in the general case. always assume that a context switch will be taken. This means that we now have only a sufficient, albeit not necessary, schedulability test. Also recall that we assume a system for ....
D. I. Katcher, J. K. Strosnider, and E. A. Hinzelman-Fortino, "Dynamic versus Fixed Priority Scheduling: A Case Study," Technical report, Dept. of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania, Aug. 1993.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC