Abstract: A request processing method and apparatus, an electronic device, and a computer storage medium are provided, which are related to the technical field of cloud computing. The request processing method includes: receiving a content delivery network (CDN) request; acquiring a dynamic code corresponding to the CDN request, wherein the dynamic code is a pre-configured code; compiling the dynamic code in real time to obtain a compiled code; and executing the compiled code. The request processing method provided in an embodiment of the present application may improve the flexibility of request processing of a CND system, and has no concurrency limitation.

Abstract: A distributed device management system specifies a device capable of supplying request data used for providing a service, from among a plurality of devices connected to a network. Device management function units are disposed so as to be geographically distributed and manage the states of the devices located in deployed areas. A device specifying function unit has a device inquiry cache in which a response log including the type of data which was previously required for the service and an identifier of the device management function unit that manages the device which was capable of supplying the data is recorded. In a case where this request data coincides with the type of data included in the response log, an inquiry is transmitted to the device management function unit associated with the request data in the response log.

Abstract: The disclosed technology relates to a process for metered training of fog nodes within the fog layer. The metered training allows the fog nodes to be continually trained within the fog layer without the need for the cloud. Furthermore, the metered training allows the fog node to operate normally as the training is performed only when spare resources are available at the fog node. The disclosed technology also relates to a process of sharing better trained machine learning models of a fog node with other similar fog nodes thereby speeding up the training process for other fog nodes within the fog layer.

Abstract: Aspects of the present invention provide an approach for exposing payloads from non-integrated machine learning systems. A generic binding identifier is established to represent a machine learning (ML) system among a set of non-integrated learning systems. A generic subscription identifier is established to represent a deployed model in the ML system. Payload data including a user request, a response, the generic binding identifier, and the generic subscription identifier are received from the ML system and stored in a database for later analysis to identify any issues related to the deployed model.

Abstract: A network monitoring method includes: monitoring, at a network node, one or more network statistics for each of one or more objects exchanged between a client end point and a server end point through a network connection, the network node being located between the client end point and the server end point; and determining one or more QoS metrics for each of the one or more objects based on the network statistics for respective one of the one or more objects.

Abstract: An information handling system may include a host system including a host network interface; and a management controller including a management network interface, the management controller being configured to provide out-of-band management of the information handling system. The management controller may be configured to: receive, via the management network interface, network configuration information for the host network interface; and provide access to the network configuration information via an application programming interface (API). The host system may be configured to: poll the API of the management controller at selected times until the management controller provides access to the network configuration information; retrieve the network configuration information via the API; and configure the host network interface in accordance with the network configuration information.

Abstract: A method includes distributing a number of printer devices across a computer network with corresponding logistical nodes, and, through a server, dynamically extracting, from a number of digital pre-print files, metadata related to documents and pre-conditions based on scanning thereof, and classifying the number of digital pre-print files into groups based on an extent to which the dynamically extracted metadata is shared between constituents thereof.

Abstract: Provided is a system and method for writing IoT data of an object in transit and business context associated with the IoT data and/or the object to a blockchain. In one example, the method may include receiving sensor data of an object that is in a multi-party transactional process, identifying dynamic context of the object based on a current position of the object within the multi-party transactional process, generating a data block comprising the received sensor data of the object and the identified dynamic context of the object, and storing the data block within a blockchain on an immutable distributed ledger.

Abstract: A computer-implemented method, a computer program product, and a computer system for efficient use of computing resources in two stage training of a deep learning model. A computer executes a first first-stage training job to train a deep learning model. The computer finishes the first first-stage training job by using early stopping and then registers a first second-stage training job to train a deep learning model that has been trained in the first first-stage training job. The computer executes the first second-stage training job with a small number of epochs. The computer interrupts the first second-stage training job and executes a second first-stage training job, in response to receiving a registration of the second first-stage training job. The computer interrupts the first second-stage training job and executes a second second-stage training job that has a higher priority, in response to receiving a registration of the second second-stage training job.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for self-optimization of computing environments. In some implementations, different combinations of settings used by one or more server environments are determined. Results achieved by the one or more server environments are monitored when using the different combinations of settings. Based on the monitored results, one or more performance measures are generated that correspond to each of the different combinations of settings. An updated set of settings are selected for a particular server environment based on the performance measures. The selected settings are provided for the particular server environment.

Abstract: A method for generating an executable program to run on a system of one or more processor chips each comprising a plurality of tiles. The method comprises: receiving a graph comprising a plurality of data nodes, compute vertices and directional edges, wherein the graph is received in a first graph format that does not specify which data nodes and vertices are allocated to which of the tiles; and generating an application programming interface, API, for converting the graph, to determine a tile-mapping allocating the data nodes and vertices amongst the tiles. The generating of the API comprises searching the graph to identify compute vertices which match any of a predetermined set of one or more compute vertex types. The API is then called to convert the graph to a second graph format that includes the tile-mapping, including the allocation by the assigned memory allocation functions.

Abstract: A method for distributing an application for an edge computing device performed by a computing device according to an embodiment of the present invention includes selecting a cluster including two or more edge devices from among a plurality of edge devices, distributing a first application to a second edge device included in the cluster, modifying routing information such that a service request incoming to a first edge device included in the cluster is transmitted to the second edge device, and replacing the first application running in the first edge device with a second application.

Abstract: A method for generating an executable program to run on a system of one or more processor chips each comprising a plurality of tiles. The method comprises: receiving a graph comprising a plurality of data nodes, compute vertices and directional edges, wherein the graph is received in a first graph format that does not specify which data nodes and vertices are allocated to which of the tiles; and generating an application programming interface, API, for converting the graph, to determine a tile-mapping allocating the data nodes and vertices amongst the tiles. The generating of the API comprises searching the graph to identify compute vertices which match any of a predetermined set of one or more compute vertex types. The API is then called to convert the graph to a second graph format that includes the tile-mapping, including the allocation by the assigned memory allocation functions.

Abstract: Some examples described herein relate to packet identification (ID) assignment for a routing network in a programmable integrated circuit (IC). In an example, a design system includes a processor and a memory coupled to the processor. The memory stores instruction code. The processor is configured to execute the instruction code to construct an interference graph based on routes of logical nets through switches in a routing network, and assign identifications to the routes comprising performing vertex coloring of vertices of the interference graph. The interference graph includes the vertices and interference edges. Each vertex represents one of the logical nets having a route. Each interference edge connects two vertices that represent corresponding two logical nets that have routes that share at least one port of a switch. The identifications correspond to values assigned to the vertices by the vertex coloring.

Abstract: Systems, methods, and devices, including computer-readable media, for managing operation of devices in complex systems and changing environments. In some implementations, a server system stores data indicating management plans for each of a plurality of different devices, each management plan indicating a device-specific set of program states for programs in a predetermined set of programs. The server system alters the management plans and enforces interdependence of the programs, and the server system generates a customized instruction that alters operation of the device according to the device-specific set of program states assigned in the altered management plan for the device. The server system causes each device to perform one or more operations of the device determined according to the device-specific set of program states assigned in the altered management plan for the device.

Abstract: Described herein are systems, apparatus, methods and computer program products configured for actions to be performed on segments within a pipeline. The disclosed systems and techniques allow a user to select a data segment on a graphical user interface. The user may then utilize the data of the segment to perform one or more actions. Thus the data of the segment may be used to provide communications to parties contained within the data, manipulate the data segment based on the parties contained within the data, and change the data displayed based on attributes selected.

Abstract: A system for controlling access to a downstream database management system (DMS) is provided. The system comprises an interface to maintain client connections with a plurality of upstream clients; a pooling component to establish a dynamic pool, and to selectively route each client connection to a pool; within each pool maintaining a queue comprising client database requests associated with particular client connections; and selectively granting access to the client database requests within each queue to at least one downstream DMS.

Abstract: Techniques for phased deployment of machine learning models are described. Customers can call a training API to initiate model training, but then must wait while the training completes before the model can be used to perform inference. Depending on the type of model, machine learning algorithm being used for training, size of the training dataset, etc. this training process may take hours or days to complete. This leads to significant downtime where inference requests cannot be served. Embodiments improve upon existing systems by providing phased deployment of custom models. For example, a simple, less accurate model, can be provided synchronously in response to a request for a custom model. At the same time, one or more machine learning models can be trained asynchronously in the background. When the machine learning model is ready for use, the customers' traffic and jobs can be transferred over to the better model.

Abstract: Techniques for heterogeneous compute instance auto-scaling with reinforcement learning (RL) are described. A user specifies a reward function that generates rewards for use with an application simulation for determining what different instance types should be added to or removed from the application as part of training a RL model. The RL model can be automatically deployed and used to monitor an application to automatically scale the application fleet using heterogenous compute instances.

Abstract: Method and apparatus for performing feature engineering using negative inferences are provided. One example method generally includes identifying a plurality of concepts and analyzing a corpus of documents to determine a first co-occurrence rate for a first concept and a second concept in the plurality of concepts. The method further includes analyzing the corpus of documents to determine a second co-occurrence rate for the second concept and at least a third concept of a set of concepts related to the first concept and determining an inverse relationship between the second concept and the third concept. The method further includes generating test data for training a machine learning model including a negative inference between the second concept and the third concept and training the machine learning model using the test data.