タイトル | Three real-time architectures - A study using reward models |
著者(英) | Smith, R. M.; Sjogren, J. A. |
著者所属(英) | Air Force Office of Scientific Research, Bolling AFB; Yale Univ. |
発行日 | 1990-01-01 |
言語 | eng |
内容記述 | Numerous applications in the area of computer system analysis can be effectively studied with Markov reward models. These models describe the evolutionary behavior of the computer system by a continuous-time Markov chain, and a reward rate is associated with each state. In reliability/availability models, upstates have reward rate 1, and down states have reward rate zero associated with them. In a combined model of performance and reliability, the reward rate of a state may be the computational capacity, or a related performance measure. Steady-state expected reward rate and expected instantaneous reward rate are clearly useful measures which can be extracted from the Markov reward model. The diversity of areas where Markov reward models may be used is illustrated with a comparative study of three examples of interest to the fault tolerant computing community. |
NASA分類 | COMPUTER OPERATIONS AND HARDWARE |
レポートNO | 91A24371 |
権利 | Copyright |
|