共查询到20条相似文献,搜索用时 78 毫秒
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
J. MacGregor 《Performance Evaluation》2003,52(4):237-267
An exact solution for the M/G/c/K model is only possible for special cases, such as exponential service, a single server, or no waiting room at all. Instead of basing the approximation on an infinite capacity queue as is often the case, an approximation based on a closed-form expression derivable from the finite capacity exponential queue is presented. Properties of the closed-form expression along with its use in approximating the blocking probability of M/G/c/K systems are discussed. Extensive experiments are provided to test and verify the efficacy of our approximate results. 相似文献
13.
14.
Roger W Hockney 《Parallel Computing》1985,2(1):1-14
We report performance measurements made on the 2-CPU CRAY X-MP at ECMWF, Reading. Vector (SIMD) performance on one CPU is interpreted by the two parameters (), and we find for dyadic operations using FORTRAN . All vector triadic operations produce ; and a triadic operation with two vectors and one scalar gives r∞ = 148 Mflop/s and . MIMD performance using both CPUs on one job is interpreted with the two parameters (), where is the amount of arithmetic that could have been done during the time taken to synchronize the two CPUs. We find, for dyadic operations using the TSKSTART and TSKWAIT synchronization primitives, that r∞ = 130 Mflop/s and . This means that a job must contain more than ~ 6000 floating-point operations if it is to run at more than 50% of the maximum performance when split between both CPUs by this method. Less expensive synchronization methods using LOCKS and EVENTS reduces to 4000 flop and 2000 flop respectively. A simplified form of LOCK synchronization written in CAL code further reduces to 220 flop. This is probably the minimum possible value for synchronization overhead on the CRAY X-MP. 相似文献
15.
Kazuko Morizawa Hiroyuki Nagasawa Noriyuki Nishiyama 《Computers & Industrial Engineering》1994,27(1-4):23-26
“Complex Random Sample Scheduling(CRSS)” was proposed in this paper as an efficient heuristic method for solving any permutation scheduling problems. To show the effectiveness of the proposed CRSS, it was applied to an N-job, M-machine, permutation flowshop scheduling problem to minimize makespan, N/M/F/Fmax. Numerical experiments made it clear that the proposed CRSS provides a schedule very close to the near-optimal schedule obtained by the existing promising heuristic methods such as taboo search and simulated annealing, within less computation time than these heuristic methods. 相似文献
16.
Consideration was given to the discrete-time queuing system with inversive servicing without interrupts, second-order geometrical
arrivals, arbitrary (discrete) distribution of the customer length, and finite buffer. Each arriving customer has length and
random volume. The total volume of the customers sojourning in the system is bounded by some value. Formulas of the stationary
state probabilities and stationary distribution of the time of customer sojourn in the system were established. 相似文献
17.
T. C. E. Cheng 《Computers & Industrial Engineering》1992,22(4):495-499
Several efficient algorithms of O(n log n) computational complexity, for the Johnson's rule to schedule a set of simultaneously available jobs on two machines in a flowship to minimize the maximum job flowtime have appeared in the literature. A modified version of one of these algorithms is presented which not only simplifies the programming effort for implementation but is also able to generate all possible optimal sequences obtainable from Johnson's rule. 相似文献
18.
H.G. Perros 《Performance Evaluation》1983,3(2):83-93
We obtain the exact analytic expression of the probability distribution of the number of units in a single server queue with Poisson arrivals and Coxian service time distribution (notated as M/Ck/1). A recursive procedure for calculating this probability distribution is given. The well-known queues M/Ek/1 and M/D/1 are re-derived as special cases of the M/Ck/1 queue. Finally, the cases of M/C2/1 and M/C3/1 are fully worked out. 相似文献
19.
20.