Edge computing can be defined as an emerging technology that uses cloud computing to leverage edge data centers to process, store, and analyze data close to the source. Traditional cloud computing architectures are not designed for latency-critical applications such as AI (Artificial Intelligence) and IoT (Internet Of Things) because they rely on low data volumes generated by applications running near highly-populated areas. When volume grows beyond 50 miles from the population center, networks experience higher latency and packet loss rates which impacts application performance. Since everyone's life is equipped with more and more IoT devices by the day, decisions should be made at a split second in edge computing. It is really crucial to perform at optimum level, some devices specially the medical wearables deal with patient life and any delay in decision making will result in disaster. Similarly, modern day autonomous self driving vehicles where late decisions can endup in accidents and really there is no room for any errors. This paper provides a new approach to improve performance of the edge computing by having two identical computing systems in which one system will act as primary and another as reserved or secondary, this system will be available at the local environment of the IoT device and not in cloud. The secondary system will be reserved for mission critical requests and whenever the primary system breaches latency threshold for response only then the request will be re-routed to the secondary system. Both the systems will sync data on background and can also serve as backup computing systems in case of any failure to one of the systems.