Abstract: The present disclosure provides an artificial intelligence-based hybrid RAID controller device. The artificial intelligence-based hybrid RAID controller device includes CPU to execute instructions to run overall operation of the artificial intelligence-based hybrid RAID controller device. In addition, the artificial intelligence-based hybrid RAID controller device includes XOR/Cipher engine module to perform encryption and decryption to provide data security. Further, the artificial intelligence-based hybrid RAID controller device includes DSP module to perform pre-processing of data for an artificial intelligence inference engine module. Furthermore, the artificial intelligence inference engine module facilitates the artificial intelligence-based hybrid RAID controller device to perform in-storage processing. Moreover, the artificial intelligence-based hybrid RAID controller device includes a plurality of PCIe controller connected to an array of SSDs.

Abstract: A protocol for providing secured IO device and storage controller handshake protocol; IO device controlled cipher settings, and secured data storage and access in memory. An IO device requesting data transfer with encryption and/or decryption, requests session keys from the processor. The processor generates a fresh public-private key pair for the session. The public key is sent to the requesting IO device; the private key is momentarily saved by the processor for the session. The requesting IO device generates a secret key and its desired cipher setting; furthermore, encrypts the secret key and cipher setting using the public key, and sends secret key and cipher setting to the processor. The processor uses the private key to decrypt the secret key and cipher setting. The cipher setting is used for configuring the data processing core. The secret key is used for encryption and/or decryption of the data being transferred. All keys are not permanently saved.

Abstract: Stacking techniques are illustrated in example embodiments of the present invention wherein semiconductor dies are mounted in a module to become a MCM which serves as the basic building block. A combination of these modules and dies in a substrate creates a package with specific function or a range of memory capacity. Several example system configurations are provided using BGA and PGA to illustrate the stacking technique. Several pin assignment and signal routing techniques are illustrated wherein internal and external signals are routed from main board to various stacked modules. Expansion can be done both on the vertical and horizontal orientations.

Abstract: Stacking techniques are illustrated in example embodiments of the present invention wherein semiconductor dies are mounted in a module to become a MCM which serves as the basic building block. A combination of these modules and dies in a substrate creates a package with specific function or a range of memory capacity. Several example system configurations are provided using BGA and PGA to illustrate the stacking technique. Several pin assignment and signal routing techniques are illustrated wherein internal and external signals are routed from main board to various stacked modules. Expansion can be done both on the vertical and horizontal orientations.

Abstract: Due to the integration of multiple I/O device controllers in a storage controller and the need to provide secure and fast data transfers between the I/O devices and the storage controller, an architecture that can perform multiple encrypt/decrypt operations simultaneously is therefore needed to service multiple transfer requests without a negative impact on the speed of transfer and processing. The present invention relates to enhancing Direct Memory Access (DMA) operations between multiple IO devices and a storage controller by adding a Data Processing Core. Exemplary implementations are provided to illustrate the background mechanism used by a DMA controller that minimizes central-processing-unit (CPU) intervention and the multi-channel architecture which allows multiple IO requests to be serviced simultaneously.

Abstract: In conventional storage device system, data transfer from memory to IO bus has to go through an intermediate volatile memory (cache). Data transfer therefore is completed in two steps—data is transferred from memory to cache and then from cache to the IO bus. Memory-to-cache transfer is handled by one DMA engine and another DMA engine for cache-to-IO transfer. To start the transfer, processor prepares the DMA transfer from memory to cache. Upon completion of the memory-to-cache transfer, the processor is interrupted to prepare the transfer from cache to IO. In between transfers, the processor has to intervene to setup the next transfer utilizing the precious processor cycles. The present invention improves on the above process using two novel schemes; 1) The use of dependency table to facilitate the transfer from memory-to-IO with less intervention from the processor and 2) the use of Bus Snooping scheme to bypass the transfer to cache making the transfer directly from memory to IO bus.

Abstract: A hybrid storage system comprising mechanical disk drive means, flash memory means, SDRAM memory means, and SRAM memory means is described. IO processor means and DMA controller means are devised to eliminate host intervention. Multi-tiered caching system and novel data structure for mapping logical address to physical address results in a configurable and scalable high performance computer data storage solution.

Abstract: Stacking techniques are illustrated in example embodiments of the present invention wherein semiconductor dies are mounted in a module to become a MCM which serves as the basic building block. Combination of these modules and dies in a substrate creates a package with specific function or a range of memory capacity. Several example system configurations are provided using BGA and PGA to illustrate the stacking technique. Several pin assignment and signal routing techniques are illustrated wherein internal and external signals are routed from main board to various stacked modules. Expansion can be done both on the vertical and horizontal orientations.