In a ring, each device has two paths to any other device, which is good for redundancy. However, all traffic passes through the ring link, which requires high bandwidth on interfaces. In addition, when the network is changed or faulty, service packets may be lost for a short time due to path switching.
To provide redundancy in a star topology, two links are required for each device. Because the bandwidth requirements of each device are independent, so the bandwidth requirements of each interface are lower than those of the ring network. However, even if links are redundant, if a core device is faulty, all services will be interrupted. Therefore, two core devices need to be deployed in active/standby mode or stack mode for redundancy.
In conclusion, the two topologies have their own advantages and disadvantages. In practice, if the cost permits, you can consider using the two topologies together.