Resource-scheduling and load-balancing algorithms in the cloud environment play a vital role in delivering various services using the internet. When any end-user submits tasks to the cloud for computations, the cloud uses these resource-scheduling and load-balancing algorithms to compute them on its virtual machines. Cloud performance is proportional to the manner in which resources and load are managed. Its performance improves with proper management of resources and load; otherwise, it outputs mediocre or low results hampering its resiliency and availability features. Depending upon its functionality, each resource scheduling algorithm differs from another; hence, to evaluate the cloud performance, it becomes crucial to study and compare these resource scheduling algorithms. The pivotal point of this study is to evaluate the cloud performance by computing tasks of various lengths using the Highest Response Ratio Next (HRRN) resource scheduling algorithm and compare its performance concerning First Come, First Serve (FCFS) and Shortest Job First (SJF). An experiment was conducted using the dataset of Alibaba task-events, where tasks were computed in multiple scenarios. The results obtained from this experiment are compared with each other using various performance metrics. Later, a detailed empirical analysis is also performed using the Linear Regression Equations mathematical model to add an extensive comparison of HRRN with FCFS and SJF. Lastly, this research paper proposes using Reinforcement Learning, a sub-domain of Machine Learning, to improve cloud performance by making its computations completely dynamic.