This paper presents an analytical method for the performability evaluation of a proposed network memory server attached to a local area network. The proposed server is expected to improve data storage and retrieval. However, the underlying network will receive an increased amount of traffic which may have a significant effect on the network?s performance. The proposed analytical model and its solution can be used to evaluate the performance of such a system. Mean queue lengths and the probability that the NMS queue is full are calculated and presented.