| dc.description.abstract |
The rapid development of technology has significantly transformed many aspects of data communication processes. Cloud computing technology, also known as cloud computing, is now widely used. With the increasing adoption of cloud services, there has been a significant increase in traffic, leading to load imbalance among the servers in the cloud environment. The objective of this research is to allocate traffic alternately to the available servers to prevent server overload and reduce the risk of system failure. Implementing load balancing using the round-robin method can improve system performance in handling high traffic in the cloud environment. Load balancing is a technique of distributing traffic load evenly across two or more connection paths to optimize traffic flow, maximize throughput, minimize response time, and avoid overload on any single path. Load balancing is employed when a server exceeds its maximum capacity with the number of users. Meanwhile, Round Robin is a process scheduling algorithm in operating systems designed to process and rotate queues alternately. Based on the test results, sending requests to the server using load balancing with 30,000 users per second resulted in an error rate of 42.37%, 50,000 users per second resulted in an error rate of 66.31%, and 100,000 users per second resulted in an error rate of 63.08% fewer than before using load balancing, where 30,000 users per second had an error rate of 62.28%, 50,000 users per second had an error rate of 75.39%, and 100,000 users per second had an error rate of 73.59%. Therefore, this research enables system scalability improvement and effective handling of increased user requests in a distributed manner.
Keyword: cloud computing, google cloud,load balancing, round robin, overload.
| en_US |
