Abstract: A memory controller is configured to be operatively connected to clients for shared memory access. The memory controller is configured to perform a recovery of a shared memory by determining starting point for the recovery and then replicating shared memory to target file system. The memory controller is configured to determine starting point by: generate journal of file operations, generate a snapshot of the shared memory; restore shared memory as per the snapshot to the target file system; determine last operation before snapshot in journal; and set next operation after last operation as starting point. The memory controller is further configured to replicate shared memory by replaying file operations according to journal starting with starting point. The present disclosure enables to create an initial consistent synchronization point of the shared memory with clients, which in turn ensures reliable data recovery of the shared memory in any adverse event.

Abstract: One example method includes receiving input concerning a mobile IoT device, and the input includes information about a location of the mobile IoT device, information about whether the mobile IoT device is moving, and, when the mobile IoT device is moving, information about the range, speed, and bearing of the mobile IoT device. Next, the method includes generating a predicted location of the mobile IoT device based on the inputs received, using the predicted location of the mobile IoT device and a map of nodes in an environment where the mobile IoT device is located to make a migration decision concerning an application used by the mobile IoT device, and migrating the application from a present location to a node expected to be accessible by the mobile IoT device when the mobile IoT device reaches the predicted location, and the node and present location are physically separated by a distance.

Abstract: A NAS memory controller is provided that is operatively connected to one or more NAS clients for shared memory access, where each NAS client includes a clock and is operatively connected to one another through a low-latency communications network. The NAS memory controller is configured to transmit an indication of a stop time and a restart time to each of the one or more NAS clients, and after the stop time arrives, receive a journal of file operations from each of the one or more NAS clients. The NAS memory controller is further configured to determine a status of each journal, and thus provides a reliable data replication by accurately synchronizing the file operations of the NAS clients using the journal without the need to have a programmatic access to the NAS source, without installation of gateways, and without requirement to reread all data written to the NAS source.

Abstract: A memory controller stores data comprising a plurality of data blocks in a first memory tier by data striping data on data drives of a first drive type. The memory controller determines a first parity for at least one data stripe of a plurality of data stripes, stores the first parity in one of data drives of the first drive type, determines a plurality of second parities for the at least one data stripe of the plurality of data stripes, and stores the second parities in a second memory tier.

Abstract: A continuous data protection unit (CDP unit) is arranged to receive from a primary splitter, a copy of incoming data sent to a primary storage in the form of incoming change sets. The CDP unit comprises CDP data mover, and CDP storage unit. The CDP data mover is arranged to receive the incoming change sets and write recovery data based on one or more change sets to the CDP storage unit and to a recovery unit arranged to hold a copy of the recovery data.

Abstract: Provided is a data processing arrangement (100, 200, 300, 400) that is coupled to a data memory arrangement (102) and is configured to generate a file catalog including information describing characteristics of data files stored within the data memory arrangement. The file catalog is periodically updated so that it provides a temporal record of the information. The data processing arrangement is configured to determine a behavioral profile (404) indicative of temporal trends or patterns in the information, and to provide a warning indication in an event that the information for a given data file temporally changes in a manner that deviates more than a threshold amount from a model of expected temporal trends or patterns of the given data file.

Abstract: Provided is a computer-implemented method of backup and restore in a network data storage system (200, 310). The method includes generating, at a predefined time interval, a series of one or more virtual snapshots including, for each item in the network data storage system: (i) a strong hash of the item data, and (ii) one or more metadata elements related to the item. The method includes, in response to a request to restore a particular item: (i) searching the series of virtual snapshots for all available sources of the particular item, (ii) identifying the best source for the particular item, and (iii) recovering item data for the particular item from the identified best source.

Abstract: A memory controller is for receiving incoming data and storing and/or transmitting the incoming data utilizing deduplication. The memory controller determines a second hash corresponding to one or more first hashes. The second hash is a sparse hash and the first hashes are strong hashes. The memory controller populates a sparse index with the second hash. The sparse index associates the second hash with the corresponding first hashes. The memory controller determines the second hash based on a probability that the second hash corresponds to a plurality of first hashes using a probabilistic method.

Abstract: Techniques are provided for reinforcement learning-based evaluation of software product usage. One method comprises obtaining key performance indicators indicating software product usage by a user; determining, for a predefined time window: (i) a mean and/or a median of the obtained KPIs; (ii) an amount of time that the software product was active; and (iii) an amount of interactions by the user with a user interface; evaluating possible login states of the software product using at least one reinforcement learning agent, wherein the evaluating comprises (a) observing the plurality of possible login states, including a current state comprising a current login state of the software product, and (b) obtaining an expected utility score for changing from the current login state to a different login state of the software product; and determining whether to change from the current login state to a different login state of the software product based on the expected utility score.

Abstract: A data protection system configured to backup a time series database is provided. The data protection system may be integrated with or have access to consolidation policies of the time series database. The backup policy and backup retention policy are set by monitoring the consolidation policy and adjusting the backup policy to ensure that the data in the time series database is protected prior to being downscaled, discarded or otherwise consolidated.

Abstract: A method of data management in a data storage system including a plurality of data storage units, the method includes dividing each file in the data storage system into a plurality of blocks having a common size. The method further includes generating a hash value for each block using a common hash algorithm. The method further includes identifying a plurality of similar files in different data storage units, based on a comparison of the hashes for each file. The method further includes copying one or more of the identified similar files, such that similar files are stored in a single data storage unit and for one or more of the copied files, generating a link to a new location of the file. The method provides an efficient, effective, and adequate reduction of the primary storage as well as the secondary storage space.

Abstract: A computer-implemented method for indexing a data item in a data storage system includes: dividing the data item into one or more large blocks; dividing each large block into one or more small blocks; calculating a strong hash value for each of the small blocks and storing a list of strong hash values with a pointer to a location of the large block; from the list of strong hash values calculated for each large block, selecting one or more representative hash values for the large block; and compiling a sparse index including an entry for each large block. Each entry is based on the representative hash values and a pointer to the list of strong hash values for each large block.

Abstract: A containerized environment and application that are configured for component specific continuous replication and granular application level application. A key value store, which stores key values related to configuration data of the containerized application, is replicated continuously to a replicated key value store at a replica site. Persistent volumes may also be replicated to a replica site. The replication can be performed to multiple replica sites in an application specific manner.

Abstract: A method of preventing duplication of encrypted data in storage arranged in network in which two or more hosts can store data. The storage includes one or more encrypted data blocks, and the network further includes a distributed ledger accessible to each of the two or more hosts. The method leverages the distributed ledger for efficient deduplication of the data even with untrusted parties, such as the hosts. The method, performed by first host, includes obtaining a first hash value, a proof of ownership, and a random key for each block in plurality of new blocks of data. The method further includes generating and storing backup set description record, storing indication of the backup set description record, and determining zero or more blocks in plurality of new blocks that are present in storage to prevent duplication.

Abstract: A computing device of a data register includes persistent storage and a data register manager. The persistent storage stores: an instance of a primary chain that specifies an organizational structure of computing devices of the data register and a secondary chain for registering data with a first portion of the computing devices. The data register manager monitors sizes of each of the portions of the computing devices specified by the organizational structure; makes a determination, based on the monitoring, that the organizational structure fails to meet at least one criteria; and in response to the determination: reorganizes the organizational structure to meet the at least one criteria by, at least in part, modifying the secondary chain.

Abstract: A method of storing data to a data storage and related device are provided, wherein data storage is divided into two or more variable size chunks, and each variable size chunks is divided into blocks, and the data storage comprises set of metadata. The data storage is also divided into two fixed size chunks. The method comprises: writing incoming write IO to a position of the corresponding size chunk in the data storage; setting the change indicator for the corresponding fixed size chunk to indicate that the corresponding fixed size chunk has been changed since it was divided into variable sized chunks. When change indicators indicate that division of the associated fixed size chunk or chunks into variable sized chunks should be changed, rechunking fixed size chunk into variable sized chunks is executed. Because of the change indicator, high deduplication is maintained even for offline deduplication in the data storage.

Abstract: A method of continuous data protection (CDP) is provided. The method includes sending, by a splitter, block I/O write data from a computing system to a primary volume and to a CDP appliance. The method further includes, recording, by a file system tracker communicatively coupled with the splitter, a log of meta data operations made to files in a file system of the primary volume and sending the log to the CDP appliance. The method further includes, recording, by the CDP appliance, the block I/O write data in a journal and a secondary volume, and recording the log of meta data operations made to files in the file system of the primary volume in a continuous catalog.

Abstract: A method of storing a data file in a key value-based storage system includes dividing the data file into at least two variable size chunks. Each of the variable size chunks is divided into blocks, where the blocks have different sizes depending on the amount of data in each block. A set of metadata is generated for the data file including information indicating the size of each block and the size of each chunk and writing the data file and the set of metadata to the key value-based storage.

Abstract: The present disclosure provides a method and system for differential deduplication of data blocks in an untrusted storage. An example system determines first data blocks of similar data blocks as reference data blocks and second data blocks of the identified similar data blocks which are compressed using the reference data blocks to enable differential deduplication of data blocks.

Abstract: An example data storage device includes a memory and a controller. The controller is configured to store at least one of the one or more data elements utilizing differential compression. The controller is further configured to receive a data element to be stored, generate a copy of the data element to be stored, and mask data to be anonymized by deleting one or more portions to be anonymized. The controller is further configured to generate similarity hashes for one or more portions of the copy of the data element with masked data for finding one or more reference portions, and compress the data element to be stored utilizing differential compression with reference to the one or more reference portions.