Abstract:
A printed circuit board assembly (PCBA) for a storage device comprising a non-volatile memory (NVM) and a multi-core processor, wherein a first core of the multi-core processor is devoted to external interface management and a second core of the multi-core processor is devoted to internal data management.
Abstract:
A control module for a multi-level data storage device having a plurality of memory devices is disclosed. The control module may include: an access determination circuit configured to determine that access has been made to a piece of data stored on at least one of the plurality of memory devices, the piece of data associated with a level being one of a first level, a second level, or a third level; a level management circuit configured to change the level from the third level to the second level or from the second level to the first level upon determining that access has been made to the piece of data; and a memory controller configured to promote the piece of data in response to whether the level is the first level, the second level or the third level, wherein at least two levels of the first level, the second level, and the third level are associated with one of the plurality of memory devices.
Abstract:
An active storage system is disclosed. The active storage system includes a storage device, a non-volatile memory and an active drive controller. The active drive controller performs data management and/or cluster management within the active storage system, the active drive controller including a data interface for receiving at least object and/or file data.
Abstract:
A data storage device is provided. The data storage device includes a data storage medium having a plurality of data blocks, a cache having a plurality of cache blocks, wherein each cache block is identified by a cache block address, a cache control memory including a memory element for each data block configured to store the cache block address of the cache block in which data of the data block is written.
Abstract:
There is provided a method of privacy-preserving logistic regression training based on homomorphically encrypted ciphertexts. The method includes: obtaining a first packed ciphertext comprising at least a portion of a first training data sample packed into a first vector of slots thereof for training a privacy-preserving logistic regression model; obtaining a second packed ciphertext comprising a plurality of weights of the privacy-preserving logistic regression model packed into a first vector of slots thereof; determining at least a first output probability of the privacy-preserving logistic regression model based on the first packed ciphertext and the second packed ciphertext; and updating the plurality of weights based on the first output probability. There is also provided a corresponding system for privacy-preserving logistic regression training based on homomorphically encrypted data.
Abstract:
A method for data reconstruction when one HOSD has failed in a cluster of Hybrid Object Storage Devices (HOSDs) is disclosed. The method includes receiving one of a read request and a write request from a server to access data from a failed one of the plurality of storage devices and reconstructing the requested data stored in the failed one of the plurality of storage devices from portions of data stored in one or more available ones of the plurality of storage devices. The method also includes sending the requested data from the reconstructed data back to the server and sending the reconstructed data to a replacement one of the plurality of storage devices. Finally, the method includes updating a reconstruction list to indicate the replacement one of the plurality of storage devices and completion of data reconstruction.
Abstract:
A method for data stripping, allocation and reconstruction in an active drive storage system including a plurality of active object storage devices, each of the plurality of active object storage devices including one or more storage devices and a controller is provided. The method includes the controller of the identified one of the plurality of active object storage devices segmenting the received data into a plurality of data chunks and generating one or more parity chunks in response to the plurality of data chunks. The method further includes the controller of the identified one of the plurality of active object storage devices reorganizing the plurality of data chunks and the one or more parity chunks in response to a number of the plurality of data chunks and a number of the one or more of the plurality of active object storage devices into which the received data is to be stored. Finally, the method includes the step of the controller of the identified one of the plurality of active object storage devices storing the plurality of data chunks and the one or more parity chunks into the one or more of the plurality of active object storage devices.
Abstract:
Described are systems and methods for implementation in relation to hard disk drives. The disclosure is direct, at least in part, to a method for writing data in a multiple actuator multiple disk system, and the multiple actuator disk system for performing that method. The method comprises receiving the data and dividing the data into at least a first predetermined portion and a second predetermined portion. The method also includes and writing the first predetermined portion of the data onto a first disk surface of the multiple actuator multiple disk system using a first actuator of the multiple actuators while writing the second predetermined portion of the data onto a second disk surface of the multiple actuator multiple disk system using a second actuator of the multiple actuators.
Abstract:
A method for defining an erasure code for system having a predetermined number of data disks is disclosed. The method includes selecting step, constructing step, determining step and repeating step. The selecting step includes selecting a predetermined acceptable number of failures for the system. The constructing step includes constructing a first Tanner graph for two failures acceptable system having predetermined number of data disks. The determining step includes determining erasure code from the first Tanner graph. The repeating step includes repeating the constructing step and the determining step by increasing the acceptable number of failures by one and constructing another Tanner graph in response to the increased acceptable number of failures by increasing number of parity nodes until the predetermined number of failures for the system is reached.
Abstract:
Data reconstruction in a RAID storage system, by determining if a parity stripe has been reconstructed and if the parity stripe has been allocated, by the checking of a reconstruction/rebuild table and a space allocation table. Before reconstruction of a parity stripe occurs, the non-volatile memory of a failed hybrid drive is checked to determine if it is accessible and if so the data is copied to the new hybrid drive instead of reconstruction occurring.