We describe the use of a stochastic algorithm, called ALOPEX, which could b
e implemented in VLSI for optimizing the buffer allocation process in ATM s
witching networks. We present the results of computer simulations for buffe
r allocation in ATM switching networks using the ALOPEX algorithm. The algo
rithm uses a scalar cost function which is a measure of global performance.
The ALOPEX works by broadcasting the global cost function to all neural pr
ocessors in the neural network. Since each neural processor solely depends
on the global cost function no interaction is needed between the neural pro
cessors and the algorithm is more amenable to massively parallel implementa
tion. The application of the ALOPEX algorithm for the buffer allocation opt
imization in ATM networks assumes limited buffer capacity. The proposed ALO
PEX-based approach takes advantage of the favorable control characteristics
of the algorithm such as high adaptability and high speed collective compu
ting power for effective buffer utilization. The proposed model uses comple
te sharing buffer allocation strategy and enhances its performance for high
traffic loads by regulating the buffer allocation process dynamically. (C)
1999 Published by Elsevier Science B.V. All rights reserved.