Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for determining a first cost and a second cost associated with a virtual memory page accessed during execution of an application, the first cost being associated with a first memory type, and the second cost being associated with a second memory type in a hybrid memory system, comparing the first cost and the second cost to provide a comparison result, determining a current location of the virtual memory page, the current location including one of the first memory type and the second memory type, and selectively migrating the virtual memory page from the current location based on the comparison result and the current location.

Abstract: Methods, systems, and computer-readable storage media for receiving, by a messaging system, a message having a key, the key indicating a tenant of a set of tenants, providing, by the messaging system, the message in a partition of a messaging queue, reading, by a service instance, the message from the partition, the service instance being in a set of services instances, each service instance executing a service of a service-based application, and in response to the message, updating, by the service instance, at least a portion of data stored within a database system, the portion of data being associated with the tenant, the database system storing data of each tenant of the set of tenants.

Abstract: Various embodiments are generally directed to a dynamic, flexible acronym decoder. A user may submit a query via one of a plurality of user interfaces. Information describing the user may be received to generate a context vector for the user. The query may be processed against a database of terms using the context vector, a machine learning model, and content tags applied to terms in the database. Processing the queries against the database may return a result set, and the ML model may be used to compute a score for each result. The results may be ordered based on the scores and returned as responsive to the query.

Abstract: Methods, systems, and computer-readable storage media for receiving queue data, the queue data including a set of products to be fulfilled by the e-commerce system, determining market share data indicating a market share for each fulfillment location in a set of fulfillment locations, providing a set of distributions, each distribution indicating a proportion of products in the set of products to be fulfilled by a respective fulfillment location of the set of fulfillment locations, processing the market share data and the queue data through a multinomial probability model to determine, for each distribution in the set of distributions, a probability, each probability indicating a percentage of order fulfillments expected to be successfully executed for a respective distribution, selecting a distribution from the set of distributions, and fulfilling orders of the set of products using the distribution.

Abstract: Data for interactions performed by users through a portal page is collected. The data comprising a plurality of sequences of interactions performed by a user on representations of products displayed in the portal page. The plurality of sequences of interactions are input to train a neural network on temporal dependencies between interactions within a sequence from the plurality of sequences. The plurality of sequences are processed by the neural network through a plurality of learning layers to generate a model for product identification. Upon receiving an interaction by a first user at the portal page in relation to a product from the products and based on the model, identifying a first product from the products to be displayed in the portal page for the first user.

Abstract: Methods, systems, and computer-readable storage media for receiving, by a messaging system, a message having a key, the key indicating a tenant of a set of tenants, providing, by the messaging system, the message in a partition of a messaging queue, reading, by a service instance, the message from the partition, the service instance being in a set of services instances, each service instance executing a service of a service-based application, and in response to the message, updating, by the service instance, at least a portion of data stored within a database system, the portion of data being associated with the tenant, the database system storing data of each tenant of the set of tenants.

Abstract: Methods, systems, and computer-readable storage media for an autoscaling framework that uses a multi-tenant interference model for selectively scaling resources in container orchestration systems based on an estimated performance, an actual performance, and a violation count of each resource unit of a plurality of resource units. In some implementations, the autoscaling framework of the present disclosure mitigates length of response times in applications composed of a set of services.

Abstract: A method and system to facilitate monitoring of vehicles, riders, and drivers that includes receiving, using a communication device, and input related to driver energy and driver availability from one or more driver devices. Further, the method may include a step of analyzing, using a processing device, the input related to driver energy to determine energy levels of the one or more drivers. Further, the method may include matching, using the processing device, the rider with a driver based on the analyzing. Further, the method may include transmitting, using the communication device, a notification to the rider device and a matched driver device.

Abstract: In an example, information about interactions by users, via the graphical user interface, with a plurality of items rendered in the graphical user interface, is obtained, the information including a timestamp of a time and date at which the corresponding interaction occurred. A series of operations is performed upon detecting a condition that triggers a ranking of items rendered in the graphical user interface, for each item for which the condition applies. A ranking score is calculated for the item and a decay is applied to the ranking score by applying a Gaussian distribution, using the timestamp of the item as an origin to the Gaussian distribution, producing a decayed ranking score. The items for which the condition applies are then ranked based on their respective decayed ranking scores and a subset is rendered in the graphical user interface in an order matching the ranking.

Abstract: Various embodiments are generally directed to a dynamic, flexible acronym decoder. A user may submit a query via one of a plurality of user interfaces. Information describing the user may be received to generate a context vector for the user. The query may be processed against a database of terms using the context vector, a machine learning model, and content tags applied to terms in the database. Processing the queries against the database may return a result set, and the ML model may be used to compute a score for each result. The results may be ordered based on the scores and returned as responsive to the query.

Abstract: Various embodiments are generally directed to a dynamic, flexible acronym decoder. A user may submit a query via one of a plurality of user interfaces. Information describing the user may be received to generate a context vector for the user. The query may be processed against a database of terms using the context vector, a machine learning model, and content tags applied to terms in the database. Processing the queries against the database may return a result set, and the ML model may be used to compute a score for each result. The results may be ordered based on the scores and returned as responsive to the query.

Abstract: Data for user interactions with products provided through a product catalog on a user interface of a web page is collected. The data for the user interactions is evaluated to determine events defined by at least one product from the products of the catalog, one or more properties of the at least one products, and user properties of a user from the users corresponding to the event. Rank values are determined for the events. A rank value defines a first probability for performing a purchase executed with a corresponding user interaction from the user interactions for a product from the catalog products performed by a first user from the users. Upon identifying user interactions from a user with the product catalog, a set of products from the products at the product catalog is determined to be provided for display based on the rank values and user properties of the user.

Abstract: Data for interactions performed by users through a portal page is collected. The data comprising a plurality of sequences of interactions performed by a user on representations of products displayed in the portal page. The plurality of sequences of interactions are input to train a neural network on temporal dependencies between interactions within a sequence from the plurality of sequences. The plurality of sequences are processed by the neural network through a plurality of learning layers to generate a model for product identification. Upon receiving an interaction by a first user at the portal page in relation to a product from the products and based on the model, identifying a first product from the products to be displayed in the portal page for the first user.

Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for receiving a query from an application, determining that the query has been received before, and in response, retrieving statistics for the query from a metadata table, providing a query execution plan (QEP) associated with the query, the QEP including a plurality of operators executed to provide a query result, one or more operators of the plurality of operators requiring input from computer-readable memory, fragmenting at least one operator of the plurality of operators into multiple fragment operators to provide a fragmented QEP (fQEP), the multiple operators being executed to each provide a portion of a result of the at least one operator, and storing the fQEP in computer-readable memory.

Abstract: Methods, systems, and computer-readable storage media for predicting a type of an event in a computer-implemented system, implementations including receiving event data including a set of features representative of an event, determining a probability for at least one feature in the set of features from a data structure that stores a plurality of feature-probability pairs, the data structure representative of a type of event, providing a joint probability based on the probability of the at least one feature, the joint probability indicating a likelihood that the event is of the type of event, comparing the joint probability to a threshold to provide a comparison, and selectively executing one or more actions based on the comparison.

Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for receiving an annotated query execution plan (aQEP), the aQEP being processed to execute a query on an in-memory database in a hybrid memory system, and including one or more annotations, each annotation indicating an output of a respective operator that is to be provided as input to a join operator, determining a payload size at least partially based on an estimated size of an intermediate output of the join operator, selecting a memory type from a plurality of memory types in the hybrid memory system based on the payload size and a cache size, and storing, after execution of the aQEP, the intermediate output on the memory type in the hybrid memory system.

Abstract: Methods, systems, and computer-readable storage media for inhibiting security threats in microservice architectures hosted on cloud infrastructures, implementations including receiving, by a microservice used in one or more microservice-based applications, a network packet including a set of features, determining, by the microservice, a probability for the set of features with respect to a set of categories, and identifying, by the microservice, that the network packet corresponds to a first category based on probabilities of the set of features, and, in response, executing an action.

Abstract: A method and system to facilitate monitoring of vehicles, riders, and drivers that includes receiving, using a communication device, and input related to driver energy and driver availability from one or more driver devices. Further, the method may include a step of analyzing, using a processing device, the input related to driver energy to determine energy levels of the one or more drivers. Further, the method may include matching, using the processing device, the rider with a driver based on the analyzing. Further, the method may include transmitting, using the communication device, a notification to the rider device and a matched driver device.

Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for determining that an object implicated in an executing application is to be allocated to memory in an in-memory system, determining a type of the object, and allocating the object to one of a first size of virtual memory page and a second size of virtual memory page of an operating system based on the type of the object.

Abstract: In an example embodiment, information about an interaction by a user, via the graphical user interface, with an item rendered in the graphical user interface, is obtained from a graphical user interface. The information includes a session identification corresponding to a networking session in which the interaction occurred and an item identification corresponding to the item interacted with. Then k hash functions are applied to a combination of the session identification and the item identification, producing k indexes. Values of bits corresponding to the k indexes in a bloom filter stored in a non-volatile memory are determined. In response to a determination that the values of bits corresponding to the k indexes are all set to one, a session counter corresponding to the item identification is incremented.