Abstract: In a data storage device, a controller can optimize predictive loading using a hardware driven data path. In one or more examples, the controller may receive a first request for data from a host system. In response to detecting a pattern, and while transferring data, corresponding to the first request, the controller may load a first portion of data for a subsequent request for data, based on the pattern, from the device memory into a buffer. The controller may also initiate a transfer of a second portion of data for the subsequent request, to a latch in the device memory. In response to receiving the subsequent request, the controller may (i) transfer the first portion of data from the buffer, and (ii) transmit a read request to the device memory, for the second portion of data, thereby causing the device memory to transfer the second portion of data.

Abstract: A controller of a storage device receives a stream of data from a host system. The stream of data corresponds to logical block addresses. The controller writes the stream of data to data block(s) in a device memory, each data block including respective super word line(s), each super word line including respective word line(s), and each word line corresponding to at least one logical block address. The controller generates a table for storing the logical block addresses in the order of data arrival. In response to receiving an update to one or more logical block addresses of the data block(s), the controller defragments at least one data block, based on the one or more logical block addresses, and writes data for one or more super word lines of the at least one data block to a new data block, based on the table, to retain the order.

Abstract: Various devices, such as storage devices or systems are configured to efficiently process and update logical mappings within control table sets. Control table sets are often groupings of logical mapping corresponding to the logical locations of data requested by a host-computing device and the physical locations of the data within the memory array. As data is written and erased, these mappings must be updated within the control table set. Received changes to these mappings are typically stored and updated in two locations: a cache memory and a control table update list. By tracking and marking various control table sets as dirty or having undergone multiple changes, additional received updates can be stored and updated in only the cache memory, bypassing the second control table change list. By only utilizing one method of updating control table sets, processing overhead is reduced and various read or write activities are more efficiently done.

Abstract: A system having a split bus flash memory and a method for operating the split bus flash memory is disclosed. The system may include a controller, a non-volatile memory (including first and second non-volatile memory chips) and the system bus. The controller is configured to communicate via an N-bit bus. The first and second non-volatile memory chips are configured to communicate via an M-bit bus, with M<N. The system bus connects the controller with the first and second non-volatile memory chips, wherein the system bus is split with some of the system bus lines connected to the first non-volatile memory chip and other of the system bus lines connected to the second non-volatile memory chip. In this way, the controller may communicate command, address and/or data with the memory chips in parallel.