Zhiguang Shan, Chuang Lin, and Yang Yang, A multiserver multiqueue network: modeling and performance analysis, J. Univ. Sci. Technol. Beijing, 9(2002), No. 5, pp. 389-395.
Cite this article as:
Zhiguang Shan, Chuang Lin, and Yang Yang, A multiserver multiqueue network: modeling and performance analysis, J. Univ. Sci. Technol. Beijing, 9(2002), No. 5, pp. 389-395.
Zhiguang Shan, Chuang Lin, and Yang Yang, A multiserver multiqueue network: modeling and performance analysis, J. Univ. Sci. Technol. Beijing, 9(2002), No. 5, pp. 389-395.
Citation:
Zhiguang Shan, Chuang Lin, and Yang Yang, A multiserver multiqueue network: modeling and performance analysis, J. Univ. Sci. Technol. Beijing, 9(2002), No. 5, pp. 389-395.
A new category of system model, multiserver multiqueue network (MSMQN), is proposed for distributed systems such as the geographically distributed Web-server clusters. A MSMQN comprises multiple multiserver multiqueue (MSMQ) nodes distributed over the network, and everynode consists of a number of servers that each contains multiple priority queues for waiting customers. An incoming request can be distributed to a waiting queue of any server in any node, according to the routing policy integrated by the node-selection policy at network-level, request-dispatching policy at node-level, and request-scheduling policy at server-level. The model is investigated using stochastic high-level Petri net (SHLPN) modeling and performance analysis techniques. Theperformance metrics concerned includes the delay time of requests in the MSMQ node and the response time perceived by the users. The numerical example shows the efficiency of the performance analysis technique.
A new category of system model, multiserver multiqueue network (MSMQN), is proposed for distributed systems such as the geographically distributed Web-server clusters. A MSMQN comprises multiple multiserver multiqueue (MSMQ) nodes distributed over the network, and everynode consists of a number of servers that each contains multiple priority queues for waiting customers. An incoming request can be distributed to a waiting queue of any server in any node, according to the routing policy integrated by the node-selection policy at network-level, request-dispatching policy at node-level, and request-scheduling policy at server-level. The model is investigated using stochastic high-level Petri net (SHLPN) modeling and performance analysis techniques. Theperformance metrics concerned includes the delay time of requests in the MSMQ node and the response time perceived by the users. The numerical example shows the efficiency of the performance analysis technique.
下载: