The present invention provides systems and methods for allocating a pool of global memory among a set of client/servers so that storage volumes associated with a plurality of client/servers are each allocated a portion of the pool of global memory for caching of data from that volume. The amount of memory to be used for caching the volume's input/output operations (I/Os), the page size, the cache type, the cache replacement policy and the maximum cache read can be specified by volume. The amount of memory to be used for caching the volume's input/output operations, the cache type, the cache replacement policy and the maximum cache read I/O size can be changed dynamically by the changing volume-based attributes.