Track: Optimization

Abstract

Managing a lot of data using a well-designed health information system can prevent medical errors and support doctors and medical service providers in providing the proper diagnosis to patients. This condition is inseparable from advances in data exchange technology commonly used, namely REST and SOAP. The REST API is beneficial in exchanging medical record data faster in real-time. Problems then arise as time goes by, where transactions significantly increase, creating a longer access time. This study solves the problem by adding design pattern technology using a caching pattern. The data retrieval process on the web service is stored in the cache so that when retrieving data, there is no need to re-query the database and then update the cache on the web service and update daily using Cron. This research uses a case study of a healthcare business intelligence system with various data taken from 3 different systems that have been integrated. Tests were carried out in on-premise and cloud server environments with variations between 5,000 and 10,000 data and tested 50 times each. After using the catching pattern, the measurement results of on-premise web services show a reduction in response time of up to 87%. For further development, an additional service is needed to manage caches that are simultaneously entered into the server so that RAM usage occurs optimally and does not cause data access time delays.