Methods and apparatuses to reduce the number of read/write operations performed by a CPU may involve duplicating source data to enable parallel processing on the source data. A memory controller may be configured to duplicate data written to a first buffer to one or more duplicate buffers that are allocated to one or more processing threads, respectively. In some implementations, the one or more duplicate buffers are dedicated buffers, and the addresses of the first buffer and the one or more duplicate buffers are stored in a register of memory controller.