Implementation and Analysis of Distributed Cache Architecture Between Virtual Machines in VMware to Reduce Memory Access Latency

Abstract

Virtualization technology allows multiple virtual machines (VMs) to run on a single physical machine, improving efficiency and flexibility. However, virtualized systems often face performance problems such as high memory access latency and repeated data requests between VMs. To address this issue, this study implements a distributed caching system using Redis as an in-memory cache shared between virtual machines. The experiment was conducted on the VMware vSphere platform using two virtual machines: one VM acted as a Redis cache server, and the other as a client for testing. Both VMs were connected using a host-only network to ensure stable communication. Testing was performed in two scenarios: without cache and with Redis cache, each executed 10 times. The main metric measured was response time in seconds. The results show a clear performance improvement after using Redis. The average response time without cache was 0.0113 seconds, while with Redis cache it decreased to 0.00046 seconds. This indicates that Redis reduced memory access latency by approximately 97.6%. The system also remained stable during testing without any connection issues. In conclusion, implementing a distributed caching architecture using Redis effectively improves response time, reduces memory access latency, and enhances system performance in a VMware virtualized environment. This study can serve as a reference for developing more efficient and responsive virtualization systems in modern computing environments.