Abstract: Threads and kernels across multiple server instances may maintain static connections to a database during operation. To apply a patch or other upgrade to the database, a notification may be sent to the database system that causes the database system to redirect new incoming connection requests for the database to a backup database. A second notification may also be sent to each of the server instances where threads maintain active connections to the database. These threads may be allowed to finish executing a current job, then when entering the idle state these threads may be directed to refresh their connections. This refresh operation may be redirected to the backup database. The patch may be applied after all connections have been switched to the backup database. This process switches static connections to the backup database as quickly as possible without interrupting existing thread operations.

Abstract: A system and method for providing transactional data privacy while maintaining data usability, including the use of different obfuscation functions for different data types to securely obfuscate the data, in real-time, while maintaining its statistical characteristics. In accordance with an embodiment, the system comprises an obfuscation process that captures data while it is being received in the form of data changes at a first or source system, selects one or more obfuscation techniques to be used with the data according to the type of data captured, and obfuscates the data, using the selected one or more obfuscation techniques, to create an obfuscated data, for use in generating a trail file containing the obfuscated data, or applying the data changes to a target or second system.

Abstract: Techniques for reducing unsafe memory access, particularly when interacting with native libraries, are disclosed. The system may receive a memory address. The system may determine that the received memory address is not associated with an existing memory segment. The system selects a particular memory segment, of a plurality of memory segments. The memory segment may have a length of zero and a size corresponding to a size of a native heap. The system may return a reference to the particular memory segment.

Abstract: The present invention relates to dimensionality reduction for machine learning (ML) models. Herein are techniques that individually rank features and combine features based on their rank to achieve an optimal combination of features that may accelerate training and/or inferencing, prevent overfitting, and/or provide insights into somewhat mysterious datasets. In an embodiment, a computer calculates, for each feature of a training dataset, a relevance score based on: a relevance scoring function, and statistics of values, of the feature, that occur in the training dataset. A rank based on relevance scores of the features is calculated for each feature. A sequence of distinct subsets of the features, based on the ranks of the features, is generated. For each distinct subset of the sequence of distinct feature subsets, a fitness score is generated based on training a machine learning (ML) model that is configured for the distinct subset.

Abstract: Techniques for suggesting filters for query terms based on previously selected query results are disclosed. Common characteristics of previously selected query results are presented as a filter. A system trains a machine learning model by obtaining historical data including query characteristics and selected query results. Based on the historical data, the system trains the machine learning model to associate the first filter field with the first search term. The system receives a first query for execution. The system applies the machine learning model to the first query to identify the first filter field as a suggestion. The system: recommends the first field for filtering a first set of search results corresponding to the first query. Responsive to receiving user input selecting a first value for the first filter field, the system filters using the first value to generate a set of filtered search results, and presents the filtered search results.

Abstract: Described herein are systems and methods for a providing Ledger as a Service (LaaS). Blockchain technology helps bring potential solutions to the distributed ledger problem, with a linear record structure to record transaction history. However, there are different types of blockchain techniques (e.g., Hyperledger, Ethereum, Quorum), and users/developers need to know the explicit features of each technique and align with the required APIs. Ledger as a Service can allow users to an develop applications more efficiently, and can allow users to easily migrate applications among different blockchain techniques and platforms (e.g., between Hyperledger and Ethereum). LaaS can also allow for simplified transactions with a blockchain, and can additionally provide simplified communication between blockchains of different types.

Abstract: Systems and methods described herein support tag based request context in a cloud infrastructure environment. Cloud administrators do not generally have the ability to restrict resource usage in existing clouds. Granting a user permission to create resources allows them to create and/or terminate any number of resources up to a predefined account limit. Tags are associated with requests for resources for allowing administrators to restrict a user's handling of resources to the appropriate level by allowing fine-tuned control of access to the resources based on the context of the request for the resources. Request context information of the request is compared against a required credential gate level for permitting handling of resources in a tenancy having the first privilege level classification, and the request is selectively granted based on the request context information matching the first required credential gate level.

Abstract: A new and improved onboarding process for configuring and activating an email message receiving system (EMRS) to accept and process email messages on behalf of a customer is disclosed. The process comprises generating a customer-account-specific Domain Name Server (DNS) name that identifies a customer of the EMRS and a mail server for accepting and processing email messages for a domain associated with the customer. The process includes providing the customer-account-specific DNS name to the customer, receiving a request to process email messages for the domain associated with the customer and identifying using the customer-account-specific DNS name, the mail server for receiving the email messages for the domain. The process further includes verifying, using the customer-account-specific DNS name, that the customer has permission for receiving the email messages for the domain. The process includes processing the email messages and delivering the messages to a destination system.

Abstract: A system and method for dynamic log management of stream processing in a distributed computing environment, such as, for example, a streaming application or stream analytics system. A streaming application can be deployed or published to a cluster, to execute as a client application. A cluster manager coordinates with worker nodes, to commit tasks associated with the streaming application. If a need arises to generate lower-level log data associated with the streaming application, for example to diagnose an underlying cause of a warning/error message, a configuration job can be committed to the cluster to execute as a separate log-configuration application. The log-configuration application operates with the cluster manager to determine the set of working nodes currently associated with the streaming application, and modify the logger configuration at those nodes, to record or otherwise provide log data according to a modified logging level, for example to provide lower-level log messages.

Abstract: In accordance with an embodiment, described herein is a system and method for autonomous firewall rule management, for use with cloud computing environments or other types of network environments. A firewall rule management automation framework provides rule management for firewalls deployed across availability domains. The system is adapted to automatically determine firewalls that can receive network traffic from a given source subnet or destination subnet; configure the firewalls with required firewall rules; monitor the firewall rules through collection of metrics snapshots and rule hit counts; and purge underused or potentially obsolete firewall rules, for example those having zero hits over a particular period of time or number of snapshots. The system provide generic support for different types of firewall devices, and autonomous management of firewall rules within large heterogeneous computer networks that may include several types of firewalls.

Abstract: Techniques are provided for selection of machine learning algorithms based on performance predictions by trained algorithm-specific regressors. In an embodiment, a computer derives meta-feature values from an inference dataset by, for each meta-feature, deriving a respective meta-feature value from the inference dataset. For each trainable algorithm and each regression meta-model that is respectively associated with the algorithm, a respective score is calculated by invoking the meta-model based on at least one of: a respective subset of meta-feature values, and/or hyperparameter values of a respective subset of hyperparameters of the algorithm. The algorithm(s) are selected based on the respective scores. Based on the inference dataset, the selected algorithm(s) may be invoked to obtain a result. In an embodiment, the trained regressors are distinctly configured artificial neural networks. In an embodiment, the trained regressors are contained within algorithm-specific ensembles.

Abstract: In accordance with an embodiment, described herein are systems and methods identifying capabilities and limitations of an orchestration based application integration. The systems and methods described herein can detect and publish/expose such a set of limitations and capabilities of the adapters/connectors to the integration platform. In this way, such metadata can be used, for example, by a user of an integration system in designing an integration of applications, or in selecting an orchestration, based upon, e.g., the published metadata that indicates the capabilities of the individual adapters/connectors, or the orchestration as a whole (or parts thereof). This way, once an orchestration is active for processing requests, then it can also publish a metadata considering the metadata of involved adapters. This metadata can be used in adjusting requests to an integration.

Abstract: A system and method is described for database split generation in a massively parallel or distributed database environment including a plurality of databases and a data warehouse layer providing data summarization and querying functionality. A database table accessor of the system obtains, from an associated client application, a query for data in a table of the data warehouse layer, wherein the query includes a user preference. The system obtains table data representative of properties of the table, and determines a splits generator in accordance with one or more of the user preference or the properties of the table. The system generates, by the selected splits generator, table splits dividing the user query into a plurality of query splits, and outputs the plurality of query splits to an associated plurality of mappers for execution by the associated plurality of mappers of each of the plurality of query splits against the table.

Abstract: Discussed herein is a routing mechanism for graphical processing units (GPUs) that are hosted on several host machines in a cloud environment. For a packet transmitted by a GPU of a host machine and received by a network device, the network device determines an incoming port-link of the network device on which the packet was received. The network device obtains a flow information associated with the packet, and computes, based on the flow information, an outgoing port-link of the network device in accordance with a hashing algorithm. The hashing algorithm is configured to hash packets received on a particular incoming port-link of the network device to be transmitted on a same outgoing port-link of the network device. The network device forwards the packet on the outgoing port-link of the network device.

Abstract: Techniques for migrating worker nodes within clusters to a new manager instance. One technique includes receiving a request to migrate or update a configuration of a cluster within a container system, where the migration or update includes switching from a first communication pathway to a second communication pathway between worker nodes and a manager instance; creating a component and associated IP address for the second communication pathway; communicating a pod specification that includes the IP address for the second communication pathway to the manager instance, where the pod specification will cause a container tool to update each of the worker nodes with the IP address for the second communication pathway; receiving a notification that all worker nodes have been updated with the IP address; and removing a component and associated IP address for the first communication pathway from the cluster.

Abstract: A computing network includes nodes of different work groups. Nodes of a work group are dedicated to transactions of the work group. If a node of a first work group is predicted to have an idleness window, a second work group may borrow the node to execute a transaction of the second work group. At least a subset of steps of the transaction may be categorized into a step group. Trees of a transaction may be categorized into one or more tree groups. A node is selected for executing a transaction, if the predicted idleness duration of the node is sufficient relative to the predicted runtime of the transaction, the step group, and/or tree group. A credit system is maintained. A first work group transfers a credit to a second work group when borrowing a node of the second work group for executing a transaction of the first work group.

Abstract: A method, a non-transitory computer readable medium, and an apparatus are provided for switching between a synchronous data entry processing and an asynchronous data entry processing for processing input into a data entry field of the GUI based on a latency threshold for validation and a latency threshold for conversion; and for determining which input from a series of inputs entered into the data field during the asynchronous data entry processing based on a criteria.

Abstract: Techniques are disclosed for automatic scene detection and character extraction. In one example, audiovisual content with video frames, an audio recording, and timing information is received. A score, based on the frame's visual characteristics, is determined for a first frame and subsequent frames. The first frame's score and subsequent frame's scores are compared to determine if the difference between the scores is above a threshold. When the difference in scores is above a threshold, the subsequent frame is classified as a new scene. The audiovisual content is segmented into scenes and textual characters are identified in at least one frame from each scene. The characters are stored and indexed in a searchable database with the timing information for the scene where the characters were identified. The audio recording is transcribed and the transcribed words are stored and indexed in the searchable database with timing information.

Abstract: A computing network includes nodes of different work groups. Nodes of a work group are dedicated to transactions of the work group. If a node of a first work group is predicted to have an idleness window, a second work group may borrow the node to execute a transaction of the second work group. At least a subset of steps of the transaction may be categorized into a step group. Trees of a transaction may be categorized into one or more tree groups. A node is selected for executing a transaction, if the predicted idleness duration of the node is sufficient relative to the predicted runtime of the transaction, the step group, and/or tree group. A credit system is maintained. A first work group transfers a credit to a second work group when borrowing a node of the second work group for executing a transaction of the first work group.

Abstract: Systems and methods for intelligently providing users with supporting information based on big-data analyses of a data set. Machine-learning algorithms may be executed using the data set to identify correlations between data objects of the data set. The correlations can be used to recommend supporting information to a user. A user interface can be provided to enable a user to initiate a process associated with an event. In response to receiving the input, the system can identify variables associated with the request. Based on these variables, the system can retrieve output data of the machine-learning algorithms to identify the supporting information for the user.