The client cache is used on both hybrid and all-flash configurations. It leverages local DRAM server memory (client cache) within the node to accelerate read performance. The amount of memory allocated is .4% -1GB per host. vSAN first tries to fulfill the read request from the local client cache, so the VM can avoid crossing the network to complete the read. If the data is unavailable in the client cache, the cache-tier SSD is queried to fulfill the read request. The client cache benefits read cache-friendly workloads.