Server virtualisation has dramatically reshaped the information technology industry in recent years. There are too many variants to consider. This paper proposes a practical method of achieving efficacious disk performance in a virtualized environment. Remote disk access in a virtualized environment is emerging as one of the mainstream network intensive applications, with large demands for disk and network bandwidth. As a performance metric we investigate the efficiency and effectiveness of using multi controllers combined with disk alignments. We investigate on how to deliver optimized storage input/output performance and availability at the lowest possible cost and present best practices that we demonstrated on our implementation of VSphere based Oracle installations. We also propose mechanisms to prevent common performance issues like interface queue saturation. Our findings suggest that an underutilized storage system can mask environment misconfigurations. Once the storage system becomes busier, these misconfigurations can cause performance issues. Similarly, storage performance issues are rarely due to a single cause, for example, we had virtual machine misalignment and background process contention. We also concluded that when we use mixed disk technologies in the same storage area network and stress one type with writes, this can also have a negative impact on the others due to the load on the storage controller. In addition to this, we also found that misalignment affects write latency more than the read operations.