Performance Analysis of Distributed and Centralized Models for Electronic Medical Record Exchanges

Electronic medical record exchanges can save time and reduce cost by eliminating redundant data and typing errors. The major steps of record exchange consist of querying information from database, encoding data into messages, and sending and decoding messages. Three medical-record-exchange models were proposed in the past, including the distributed, centralized, and indexed models. In this paper, the queuing theory is applied to evaluate the performance of the three models. We estimate the service time for each queue of the CPU, database and network, and predict the response time, probable bottlenecks and system capacities of each model.