Web applications today, are widely used by businesses to reach global audience. Ecommerce and other commercial applications that leverage business performance need to be given highest priority since their performance leads success. Unplanned downtime and server connection timeout error are some of the most prominent forms of failures in web service experienced by the web users, which could be as a result of network disruption or congestion of loads on the web server. This work adopted Service Worker (SW) caching technique to create a fault-tolerant web application that maintained page uptime in the event of network failure and prevented server time-out error. Two demRecords:b Application modules of Electronic Health Records; SW1 module and SW2 module of the same quality and contents were acquired from CLINIC365 domain. The Service Worker caching technique was designed with the four stages of JavaScript procedures which include SW registration, SW installation, SW return request, and SW cache updates. The designed SW caching technique was implemented in SW1 module. The performance of the acquired SW1 and SW2 modules were evaluated and compared using uptime, response time, and throughput. Throughput and response time were collected using GTMETRIX web analytic software, degree of uptimes were collected by establishing a network test with Google console tool network throttle. The performance of fault tolerant SW1 yielded
uptime, response time, and throughput of 4 points, 2.6 seconds, and 16.2 Mbps. respectively, while SW2 yielded 0 point, 4.3 seconds, and 12.6 Mbps. This study showed that there was throughput gain with SW1 module, and its uptime was maintained compared to SW2 module. In addition, response time in SW1 was faster compared to that of SW2
which is a great advantage in the prevention of server congestion, and connection time-out errors.