Optimization Method Based on Data Migration on Metaverse

As one of the most notable frameworks in the realm of Metaverse data stream computing, Flink struggles with rescaling cluster resources amidst persistent fluctuations in processing loads, which leads to performance bottlenecks. To tackle this issue, the method of data migration has been introduced. Additionally, a resource-constrained model and an optimal data migration model have been developed to establish a theoretical foundation, paving the way for proposing a data migration-based method for constructing Metaverse data streaming clusters in Flink. This method consists of a resource constraint algorithm and a data migration algorithm. The resource constraint algorithm assesses the appropriateness of a cluster for data migration by estimating the utilization levels of CPU, memory, and network bandwidth across each work node. Based on the resource constraint model, the data migration algorithm then designs an optimal method for migrating data. Experimental results demonstrate that, when deployed on a heterogeneous cluster using standard benchmarks, the proposed method effectively reduces latency and enhances the reliability of data processing.