基于Fusion Compute虚拟化平台的HIRFL数据库迁移及性能优化

为解决兰州重离子加速器（HIRFL）数据中心资源利用率低下、硬件和维护成本较高等问题，HIRFL数据中心正在向虚拟化集群模式进行迁移。本文以华为Fusion Compute虚拟化平台为基础，构建了HIRFL数据库的迁移方案。通过物理主机、虚拟机计算资源、内存、存储和网络资源性能优化，提出了1套完整的HIRFL数据库虚拟机性能优化方案。采用业界主流的数据库性能测试工具Benchmark Factory对迁移前后的数据库进行了测试，对比实验表明，本文提出的方案具有有效性和可行性，降低了数据中心的运行和维护成本，提升了HIRFL数据库系统的安全性和稳定性。

HIRFL Database Migration and Performance Optimization Based on Fusion Compute Virtualization Platform

In order to solve the problems of low resource utilization and high hardware and maintenance costs of the Lanzhou Heavy Ion Accelerator (HIRFL) data center, the HIRFL data center is moving to the virtualized cluster mode. In this paper, a migration plan for HIRFL database was built based on Huawei’s Fusion Compute virtualization platform. A complete set of performance optimization schemes for HIRFL database virtual machine was proposed, which optimized the performance of physical host, computing resources, memory, storage and network resources. Benchmark Factory was used to test the database performance before and after migration. The comparison experiments show that the proposed scheme is effective and feasible, reduces the operation and maintenance cost of the data center, and improves the security and stability of the HIRFL database system.