Abstract: The disclosure relates to methods and systems for predicting time of occurrence of future server failures using server logs and a stream of numeric time-series data occurred with a particular time window. Method performed by processor includes accessing plurality of server logs and stream of numeric time-series data, applying density and sequential machine learning model over plurality of server logs for obtaining first and second outputs, respectively, applying a stochastic recurrent neural network model over the stream of time-series data to obtain third output. The method includes aggregating first, second, and third outputs using an ensemble model, predicting likelihood of at least one future server anomaly based on the aggregating, and determining time of occurrence of the at least one future server anomaly by capturing server behavior characteristics using time-series network model. The server behavior characteristics include time-series patterns of the stream of numeric time-series data.

Abstract: Systems and methods for predicting whether a nonvolatile memory block is likely capable of being securely erased to be eligible for composing into another composable infrastructure are described. A management module receives a secure-erase command to erase at least one nonvolatile memory block, determines health parameters of the nonvolatile memory block, calculates a failure index based on the health parameters, and, based on the failure index, either securely erases the block of memory or retires the nonvolatile memory block.

Abstract: Systems and methods for data storage device firmware updates in a composable infrastructure are described. Recomposition of a system in a composable infrastructure may include downloading a firmware update to a data storage device, activating the firmware update in the data storage device, and receiving a reset command at the data storage device. The reset command may include a step to flush a virtual-to-physical (V2P) mapping table from a volatile memory in the data storage device to non-volatile memory devices in the data storage device. Responsive to a flag that is set and the received reset command, the data storage device may forgo the step of flushing the V2P mapping table.

Abstract: Methods, systems and computer program products implementing hierarchical classification techniques are disclosed. A hierarchical classification system receives training data including labeled transaction records. The system determines tag sequences from the training data. The system clusters the tag sequences into clusters. The system determines a cluster-level classifier that is trained to predict a cluster for an input transaction record. The system determines a respective cluster-specific classifier for each cluster. The system trains the cluster-specific classifier to predict a label of entity of interest for an input transaction record, given a particular cluster. Upon receiving a test transaction record, the system first applies the cluster-level classifier to determine a particular cluster for the test transaction record, and then determines a label of entity of interest of the test transaction record by applying a cluster-specific classifier of that particular cluster.

Abstract: Systems and methods for managing queue limit overflow for data storage device arrays are described. Host storage connections are allocated by host connection identifier and storage device processing queues are allocated by completion connection identifier through a connection virtualization layer. Storage commands may be directed to a processing queue based on the host connection identifier. Responsive to determining that the processing queue has reached its queue depth limit, another processing queue is determined for receiving the storage command without indicating processing queue overflow to the host device.

Abstract: Systems and methods for connection virtualization in data storage device arrays are described. A host connection identifier may be determined for a storage connection request. A target storage device and corresponding completion connection identifier may be determined for a storage command including the host connection identifier. A command tracker may be stored that associates the storage command with the host connection identifier and the completion connection identifier and the storage command may be sent to the processing queue associated with the completion connection identifier.

Abstract: Systems and methods for capped allocation of storage resources based on connection type are described. A new storage connection request from a host system to a group of data storage devices includes a connection type, such as a remote direct memory access connection or transmission control protocol connection. Based on the connection type, a subset of a pool of allocated storage resources for that connection type is determined. If the new connection request does not exceed the maximum number of connections allocated to the subset, a new storage connection is established. If the new connection request exceeds the maximum number of connections allocated to the subset, the new storage connection request fails.

Abstract: Systems and methods for managing queue limit overflow for data storage device arrays are described. Host storage connections are allocated by host connection identifier and storage device processing queues are allocated by completion connection identifier through a connection virtualization layer. Storage commands may be directed to a processing queue based on the host connection identifier. Responsive to determining that the processing queue has reached its queue depth limit, another processing queue is determined for receiving the storage command without indicating processing queue overflow to the host device.

Abstract: Embodiments provide artificial intelligence-based methods and systems for predicting account-level risk scores associated with cardholders. Method performed by server system includes accessing payment transaction data and cardholder risk data associated with cardholder. The payment transaction data includes transaction variables associated with past payment transactions performed at Point of Interaction (POI) terminals within a particular time window. Method includes generating cardholder profile data based on the transaction variables and the cardholder risk data. Method includes determining account-level risk scores associated with the cardholder based on cardholder profile data. Each account-level risk score of account-level risk scores is determined by a trained machine learning model. The account-level risk scores include a wallet reload risk score, an account reissuance risk score, and a transaction channel risk score.

Abstract: Methods, systems and computer program products generating diverse and representative set of samples from a large amount of transaction data are disclosed. A data sampling system receives transaction records. Each transaction record has multiple text segments. The system selects a subset of transaction records that contain least frequently appeared text segments. The system determines a respective vector representation for each selected transaction record. The system can measure similarity between transaction records based on the vector representations. The system assigns the selected transaction records to multiple clusters based on the vector representations and designated dimensions of importance. The system identifies one or more anchors that include transaction records on boundaries between clusters. The system filters the subset of transaction records by removing transaction records that are close to the anchors.

Abstract: Systems and methods for capped allocation of storage resources based on connection type are described. A new storage connection request from a host system to a group of data storage devices includes a connection type, such as a remote direct memory access connection or transmission control protocol connection. Based on the connection type, a subset of a pool of allocated storage resources for that connection type is determined. If the new connection request does not exceed the maximum number of connections allocated to the subset, a new storage connection is established. If the new connection request exceeds the maximum number of connections allocated to the subset, the new storage connection request fails.

Abstract: Systems and methods for failover port forwarding between peer storage nodes are described. Storage nodes may include separate data ports for host network communication and peer network communication. In the event of host port failure, host nodes may be configured to send failover storage requests to a different storage node and that storage node may forward the failover storage request through the peer ports to reach the target storage node.

Abstract: Methods, systems and computer program products implementing hierarchical classification techniques are disclosed. A hierarchical classification system receives training data including labeled transaction records. The system determines tag sequences from the training data. The system clusters the tag sequences into clusters. The system determines a cluster-level classifier that is trained to predict a cluster for an input transaction record. The system determines a respective cluster-specific classifier for each cluster. The system trains the cluster-specific classifier to predict a label of entity of interest for an input transaction record, given a particular cluster. Upon receiving a test transaction record, the system first applies the cluster-level classifier to determine a particular cluster for the test transaction record, and then determines a label of entity of interest of the test transaction record by applying a cluster-specific classifier of that particular cluster.

Abstract: Disclosed herein is a fuel for use in a combustor of a gas turbine, wherein the fuel is a gas mixture that comprises hydrogen and exhaust gas from a total combustor.

Abstract: The disclosure relates to methods and systems for predicting time of occurrence of future server failures using server logs and a stream of numeric time-series data occurred with a particular time window. Method performed by processor includes accessing plurality of server logs and stream of numeric time-series data, applying density and sequential machine learning model over plurality of server logs for obtaining first and second outputs, respectively, applying a stochastic recurrent neural network model over the stream of time-series data to obtain third output. The method includes aggregating first, second, and third outputs using an ensemble model, predicting likelihood of at least one future server anomaly based on the aggregating, and determining time of occurrence of the at least one future server anomaly by capturing server behavior characteristics using time-series network model. The server behavior characteristics include time-series patterns of the stream of numeric time-series data.

Abstract: The present invention relates to processes for the preparation of biofuel from biomass by fast hydropyrolysis or fast pyrolysis, using hydrogen generated by sorption enhanced steam reforming. The present invention also relates to fixed bed tandem catalytic-upgrading processes, and reactors and hydrodeoxygenation (HDO) catalysts useful in those processes.

Abstract: Systems and methods for distributed storage systems using synchronous discovery logs for fabric subsystem discovery. Storage nodes may be configured with fabric services and a management service may provide synchronization of discovery logs across fabric subsystems and between peer fabric nodes. The peer fabric nodes may communicate with one another using a fabric network protocol and corresponding fabric subsystems may communicate with each peer fabric node. When a target subsystem fabric node updates its discovery log, the updated discovery log is sent to the corresponding peer fabric node. The corresponding peer fabric node sends the updated discovery log to each subsystem fabric node and to each peer fabric node for synchronization across all subsystem fabric nodes. A host may contact any subsystem fabric node for the updated discovery log.

Abstract: Disclosed herein is a gas capture system comprising: a gas inlet arranged to receive a gas flow into the system; a gas outlet arranged to provide a gas flow out of the system; a gas capture region for mass transfer between a gas and a sorbent of the gas; and a sorbent regeneration region for regenerating the sorbent by heating the sorbent so that the sorbent releases a gas; wherein: the gas capture region is arranged to receive sorbent from the sorbent regeneration region; the sorbent regeneration region is arranged to receive sorbent for regeneration from the gas capture region; the sorbent is a solid sorbent of carbon dioxide gas; and the gas capture region comprises: a sorbent inlet arranged to receive an input of sorbent into the gas capture region; a sorbent outlet arranged to provide an output of sorbent from the gas capture region; one or more mass transfer regions arranged between the sorbent inlet and the sorbent outlet such that, in use, the sorbent is retained within the one or more mass transfer r

Abstract: Disclosed herein is a method of regenerating a sorbent of gas in a capture process of said gas, wherein the capture process comprises recirculating the sorbent between a gas capturing system and regenerating reactor system, the method comprising the regenerating reactor system performing the steps of: receiving a solid sorbent to be regenerated, wherein the sorbent is a sorbent of carbon dioxide gas; generating heat by combusting a fuel with an oxidising agent in the presence of a catalyst; regenerating the sorbent by using the generated heat to indirectly heat the sorbent so that the sorbent releases carbon dioxide gas; outputting the regenerated sorbent; and outputting the released carbon dioxide gas. Advantages of the gas capture system include a higher efficiency than known techniques.

Abstract: A system includes a host interface, a storage interface, and one or more control circuits coupled to the host interface and coupled to the storage interface. The one or more control circuits include a common set of registers configured to maintain first entries according to a first storage protocol for first storage devices connected to the storage interface and to maintain second entries according to a second storage protocol for second storage devices connected to the storage interface.