Abstract: A search engine responding to a user query to find relevant data assets in a federation business data lake (FBDL) system based on interactions of known users interacting with data assets in the FBDL system. Predicted data usage for a population of possible users is derived by training a generative model that uses reconstructive self-supervised learning (SSL) techniques to generate possible values for missing data usage features of the possible users. The predicted usage is then used to generate similarity scores that are combined with those of the known users to return relevant results to a target user. The predicted usage for possible users is processed in a consensus mechanism to produce integrated recommendations that are iteratively optimized using a chosen usage by the target user an a optimizer using a genetic algorithm to change consensus parameters to derive a final recommendation to the target user.

Abstract: A search engine responding to a user query to find relevant data assets in a federation business data lake (FBDL) system based on interactions of known users interacting with data assets in the FBDL system. Data assets are optimally placed for minimal latency or maximal load. Data asset recommendations and past data asset access information are input as features to a time-series model for predicting future data access patterns. An expected latency and load risk is then determined and scored by a weighted mean of these values, and placement optimization is simulated using an optimization method (e.g., genetic algorithm). Using the scoring and simulation, a data asset placement engine is then used to move the locations of the data assets to minimize maximal load that comprises a load risk representing how close a current load is to a service level agreement (SLA) requirement set by a system provider.

Abstract: Techniques are provided for detecting anomalous quantities in data records using predicted quantities of similar entities.

Abstract: Techniques are provided for data record anomaly reconciliation using machine learning models. One method comprises obtaining a data record comprising multiple line items; assigning the line items to a given cluster of similar line items to determine a line item neighborhood score for each line item based on a comparison of a given line item to other available line items in the assigned cluster; applying features of the data record to a machine learning model to determine a data record score for the data record based on a combination of the line item neighborhood scores for the data record; identifying anomalies in the data record based on the data record score and/or the line item neighborhood scores for the data record; and adjusting parameters of the line items to address the anomalies identified in the data record to produce a reconciled data record, based on the line item neighborhood scores and/or predefined adjustment rules.

Abstract: Techniques are used for balancing load on a storage system according to multiple variables. The techniques may be used to provide, among other things, defining, across at least two variables, a balance constraint for a load on a storage system. Among a set of transfers of volumes from one node to another node in the storage system, a transfer of a volume that minimizes the distance between the load and an ideal balanced state of the storage system is identified. The identified transfer of a volume is added to a combination of transfers of volumes. Whether the combination of transfers of volumes meets the balance constraint is determined. If the combination meets the balance constraint, the combination is selected as a solution to balance the load.

Abstract: Identifying risky code in microservices is described. A system identifies a modification to code segments in an application. The system determines a segments similarity score based on comparing the code segments, which are in a microservice of the application, against other code segments, which are in another microservice of the application. The system determines whether the segments similarity score satisfies a segments similarity threshold. The system outputs a recommendation to make the modification, which was made to the code segments in the microservice of the application, to the other code segments in the other microservice in the application, in response to a determination that the segment similarity score satisfies the similarity threshold.

Abstract: Techniques are used for balancing load on a storage system according to multiple variables. The techniques may be used to provide, among other things, defining, across at least two variables, a balance constraint for a load on a storage system. Among a set of transfers of volumes from one node to another node in the storage system, a transfer of a volume that minimizes the distance between the load and an ideal balanced state of the storage system is identified. The identified transfer of a volume is added to a combination of transfers of volumes. Whether the combination of transfers of volumes meets the balance constraint is determined. If the combination meets the balance constraint, the combination is selected as a solution to balance the load.

Abstract: Techniques are provided for detecting malicious software code embedded in image files, using machine learning. One method comprises obtaining metadata for an image file; applying the obtained metadata to at least one machine learning technique to classify the image file into at least one of a plurality of predefined classes, wherein the plurality of predefined classes comprises at least one malicious file class; and determining whether the image file comprises malicious software code based on the classification. The machine learning technique can be trained using image files classified into at least one of the plurality of predefined classes. The machine learning technique may employ a deep neural network and/or a convolutional neural network to classify the image file into the at least one predefined class.

Abstract: Techniques are provided for recommending hardware configuration changes using a user satisfaction rating. One method comprises obtaining usage data indicating user activity for users on computing devices; generating a user profile for each user; clustering the users into user clusters based on the user profiles; determining, for a given user cluster, a satisfaction score for each user in the given user cluster based on the obtained usage data for each user on the computing device; providing suggested hardware upgrades for the computing device of a given user in the given user cluster, wherein the given user is selected based on a lower corresponding satisfaction score relative to the satisfaction scores of other users in the given cluster, and wherein the one or more suggested hardware upgrades are based on hardware configurations of other users in the given cluster having a higher corresponding satisfaction score.

Abstract: Systems and methods for performing data protection operations including replication operations. A replication operation may automatically learn and predict when a replication system will need to switch modes, such as to a protective mode or to a fast-forward mode. The replication operation ensures that the data is replicated in a manner that optimizes the ability to retain data needed to perform point in time recovery operations while prioritizing the replication operation of new data.

Abstract: Techniques are provided for mapping storage volumes to storage processing nodes in a storage system. One method comprises determining a number of input/output operations associated with each of multiple storage volumes, wherein the input/output operations associated with a given storage volume are processed by a corresponding storage processing node based on a mapping of the storage volumes to the storage processing nodes; obtaining constraints that limit the input/output operations processed by at least a subset of the storage processing nodes; identifying at least one possible mapping of the storage volumes to the storage processing nodes that satisfy the constraints; and selecting an alternative mapping of the storage volumes to the storage processing nodes by applying a designated function to the at least one possible mapping. The input/output operations may comprise read operations and write operations, and the read and write operations may be balanced separately.

Abstract: A customer blockchain data store is provided. An exemplary method comprises obtaining a blockchain associated with a given customer of an enterprise having multiple customer communication channels, wherein the blockchain comprises transaction data for the given customer with the customer communication channels; obtaining new transaction data for the given customer for a given one of the customer communication channels; providing the new transaction data for the given customer to additional customer communication channels; receiving a validation of the new transaction data from the additional customer communication channels based on one or more predefined validation criteria; and storing the validated new transaction data for the given customer in the blockchain associated with the given customer.

Abstract: One example method includes identifying a changed microservice whose code has been changed, mapping, for each microservice in a group of microservices that includes the changed microservice, microservice dependencies associated with the changed microservice, determining, for each microservice in the group, a relative risk that the microservice will be adversely affected by the change to the code of the changed microservice, and based on the respective relative risks, generating a test order indicating an order in which the microservices in the group will be tested.

Abstract: One example method includes collecting telemetry data for each of a group of virtual machines (VM), and each of the VMs is associated with a user, collecting usage data for each of the VMs, creating a user profile definition for each user, and the user profile definition is created based on the telemetry data and usage data of the VMs associated with that user, creating, for each user, a user profile that is based on the user profile definition for that user, clustering the users based on similarity of their respective user profiles, and generating a recommended VM hardware configuration for a VM of one of the users.

Abstract: A method comprising: retrieving one or more customer engagement records, the one or more customer engagement records being associated with an opportunity for making a sale to a customer; identifying a plurality of free text samples that are part of the customer engagement records; identifying a plurality of data items that are part of the free text samples; calculating a plurality of semantic distances, each of the semantic distances corresponding to a different one of a plurality of data item pairs; clustering the data items into textual clusters based on the identified semantic distances; and training a classifier based, at least in part, on the textual clusters, the classifier being configured to receive an offer for the customer, classify the offer and output an estimation of whether the offer is expected to result in a sale.

Abstract: A search engine responding to a user query to find relevant data assets in a federation business data lake (FBDL) system by monitoring and recording all of the interactions of users interacting with data assets in the FBDL system, providing all of the user interactions to a recommendation engine, calculating relevance of information in the FBDL system to each user, and recommending one or more new data assets to a target user based on the relevance of the information. The relevance comprises the target user's past interactions with the data assets based and the cumulative interactions of other users with the data assets, such that if one or more of the other users has similar interaction behavior to the target user, then knowledge of the one or more other users can impact the relevance of the information with regard to the one or more new data assets suggested to the target user.

Abstract: One example method includes collecting telemetry data for each of a group of virtual machines (VM), and each of the VMs is associated with a user, collecting usage data for each of the VMs, creating a user profile definition for each user, and the user profile definition is created based on the telemetry data and usage data of the VMs associated with that user, creating, for each user, a user profile that is based on the user profile definition for that user, clustering the users based on similarity of their respective user profiles, and generating a recommended VM hardware configuration for a VM of one of the users.

Abstract: Systems and methods for detecting cost anomalies in a data protection system. Data is collected for assets of a data protection system operating in a cloud. The data often relates to cost and may constitute time series. The time series are then analyzed by performing a fitting competition using multiple models. The best fitting model is selected and the residuals are analyzes to find outliers and produce a normal zone for the signal. The outliers can identify cost anomalies that may reflect the health of the data protection system.

Abstract: Techniques are provided for facial recognition using a high probability group database. One method comprises maintaining (i) a first database of facial images of individuals, and (ii) a second database of facial images comprising a subset of the individuals from the first database based on a probability of individuals appearing in sequences of image frames at a given time; applying a face detection algorithm to sequences of image frames to identify one or more faces in the sequences of images; and applying a facial recognition to at least one sequence of image frames using at least the second database to identify one or more individuals in the at least one sequence of image frames. The second database is comprised of facial images of: (i) individuals from multiple angles; (ii) individuals that appeared in prior image frames; and/or (iii) individuals that appeared in an image frame generated by a plurality of cameras.

Abstract: Techniques are provided for optimizing hybrid cloud usage. In an example, a cloud spot manager can manage spot virtual machine instances for on-premises systems for multiple different customers. Where a customer requires more resources on its system, the cloud spot manager can terminate another customer's spot virtual machine on that system. Where a customer needs more resources than can be provided by its system, the cloud spot manager can determine another customer system to locate the first customer's spot virtual machine, and instantiate that virtual machine there.