This paper describes an end-to-end parallel communications scheme base
d on a vector routing algorithm (VRA) for ATM network control and mana
gement. An information string is partitioned into m parts, which are t
hen coded into k > m parts and sent out on k separate subchannels to t
he receiver. When tit of the k parts are received correctly, the origi
nal information can be reconstructed. Two desirable effects are achiev
ed in the context of ATM traffic control: (1) the burstiness of the so
urce traffic can be smoothed out by the partition process; (2) the qua
lity of service in terms of error, loss and delay can be controlled us
ing the number of redundant routes k - m as a control parameter. Our r
esults show that VRA is especially suitable for services with highly b
ursty traffic. We argue that several network management issues, includ
ing reliability, evolution and integration, security, and administrati
on and billing can be addressed in a simple manner using the VRA frame
work. (C) 1997 Published by Elsevier Science B.V.