Improving Task Scheduling on The Cloud Through Low Latency and Cost Effective Technique

Abstract

Offloading is emerging as a promising idea to allow handheld devices to access intensive applications without incurring performance or power costs. This is especially useful for businesses that want to run line-of-business applications on handhelds.. However, developing applications using cloud computing resources is a challenge because supporting the low-latency and scalability needs of applications requires highly dynamic orchestration of heterogeneous resources at different levels of the network hierarchy. It Is difficult. To reduce this complexity, this application provides a programming model that provides simplified programming abstractions and supports applications that scale dynamically at runtime. The goal of offloading is to decide for or against offloading. Code offload can make decisions in a number of ways. Much of the research on code swapping focuses on more sophisticated if-else conditions. Code offloading machine learning is currently an important research topic. Machine learning methods are employed to significantly improve the currently recommended methods and to perform robustly when tackling a wide range of learning tasks