Dell Products L.P. patent applications published on December 14th, 2023

Patent applications for Dell Products L.P. on December 14th, 2023

DOCKED COOLING CONTROL FOR A HANDHELD INFORMATION HANDLING SYSTEM (17749416)

Main Inventor

Robert C. Brooks, JR.

Brief explanation

An information handling system has a storage that stores a thermal table associated with the system. When the system is connected to a dock, a processor sends a temperature request to the dock. The processor receives temperature values for both the dock and the system. It retrieves thermal table data and generates a fan control signal based on the temperature values and the thermal table data, which is then provided to the dock.

• The patent application describes an information handling system that can control the fan speed based on temperature values.
• The system includes a storage that stores a thermal table, which contains data related to temperature control.
• When the system is connected to a dock, a processor sends a temperature request to the dock to obtain temperature values for both the dock and the system.
• The processor then retrieves the necessary thermal table data from the storage.
• Based on the temperature values and the thermal table data, the processor generates a fan control signal.
• The fan control signal is then provided to the dock, allowing it to adjust the fan speed accordingly.

Potential Applications

• This technology can be applied in various information handling systems, such as laptops, desktop computers, or servers.
• It can be particularly useful in docking stations or docking solutions where the system is connected to external peripherals or displays.

Problems Solved

• The technology solves the problem of inefficient fan control in information handling systems.
• By dynamically adjusting the fan speed based on temperature values and thermal table data, the system can effectively manage heat dissipation and prevent overheating.

Benefits

• The technology helps to optimize the performance and lifespan of information handling systems by ensuring proper cooling.
• It can enhance user experience by reducing noise levels when the system is connected to a dock.
• The dynamic fan control based on temperature values and thermal table data improves energy efficiency by only running the fan at necessary speeds.

Abstract

An information handling system includes a storage that stores a thermal table associated with the information handling system. A processor detects that the information handling system is connected to a dock. In response to the information handling system being connected to the dock, the processor provides a dock temperature request to the dock. The processor receives a first temperature value for the dock, and receives a second temperature value for the information handling system. The processor retrieves thermal table data from the thermal table. The processor generates a first fan control signal based on the first and second temperature values and the thermal table data, and provides the first fan control signal to the dock.

INFORMATION HANDLING SYSTEM MOUSE POWER MANAGEMENT (17839990)

Main Inventor

Peng Lip Goh

Brief explanation

The abstract describes a mouse for an information handling system that has a position sensor and a push button. The position sensor can detect movement with either high sensitivity (higher power consumption) or low sensitivity (lower power consumption). The push button can detect button press inputs with either high scan rate (higher power consumption) or low scan rate (lower power consumption). The mouse can switch between different power consumption modes based on usage patterns or commands from the information handling system.

• The mouse has a position sensor that can detect movement with high or low sensitivity.
• The mouse has a push button that can detect button press inputs with high or low scan rate.
• The mouse can switch between high and low power consumption modes based on usage patterns or commands from the information handling system.

Potential Applications

• This technology can be used in various information handling systems that require a mouse input device.
• It can be beneficial for applications that require precise movement detection, such as graphic design or gaming.
• It can also be useful for applications that prioritize power efficiency, such as laptops or mobile devices.

Problems Solved

• The mouse provides flexibility in sensitivity and power consumption, allowing users to adapt it to different applications.
• It solves the problem of having a fixed sensitivity or power consumption level, which may not be suitable for all use cases.
• It addresses the need for a mouse that can conserve power when not in use or when power efficiency is a priority.

Benefits

• Users can adjust the sensitivity and power consumption of the mouse based on their preferences or specific application requirements.
• The mouse can conserve power when not in use or when power efficiency is important, extending battery life.
• It provides a more versatile and adaptable mouse experience for different types of users and applications.

Abstract

An information handling system mouse includes a position sensor that detects movement and a push button that detects button press inputs for communication to an information handling system. The position sensor detects positions with a high sensitivity having a higher power consumption or a low sensitivity having a lower sensitivity. The push button detects button press inputs with a high scan rate having a higher power consumption or a low scan rate having a lower power consumption. The mouse configures to plural high and low power consumption modes based upon a detected usage pattern and/or a command from an information handling system, such as to adapt the mouse to use by different applications executing on the information handling system.

INFORMATION HANDLING SYSTEM MOUSE ADAPTIVE RESOLUTION (17839974)

Main Inventor

Peng Lip Goh

Brief explanation

The abstract describes an information handling system mouse that automatically adjusts its position resolution based on the display resolution, ensuring consistent cursor movement across different displays.

• The mouse has a position sensor that detects movement and generates cursor movement on multiple displays.
• It automatically switches between two position resolutions to match the display resolution where the cursor is presented.
• This ensures that cursor movement remains consistent even when the cursor moves between displays with different resolutions.

Potential Applications

• This technology can be used in any information handling system that utilizes multiple displays.
• It can be particularly useful in multi-monitor setups or systems with displays of varying resolutions.

Problems Solved

• In systems with multiple displays, cursor movement may not be consistent due to differences in display resolutions.
• This technology solves the problem by automatically adjusting the mouse's position resolution to match the display resolution, ensuring consistent cursor movement.

Benefits

• Consistent cursor movement across displays with different resolutions improves user experience and productivity.
• Users do not need to manually adjust the mouse settings when moving between displays with different resolutions.
• The technology simplifies the setup and use of multi-monitor systems.

Abstract

An information handling system mouse includes a position sensor that detects movement of the mouse and reports the movement to generate cursor movement at plural displays interfaced with the information handling system. The mouse automatically changes between a first position resolution and a second position resolution to generate cursor movement based upon a resolution of a display at which the cursor is presented. As the cursor moves between each of plural displays in response to mouse movement, the mouse automatically changes the position resolution so that cursor movement relative to the mouse movement remains consistent even where the cursor is presented at displays having different display resolutions.

METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR LATENCY PROCESSING (17874070)

Main Inventor

Bing LIU

Brief explanation

The present disclosure describes a method, device, and computer program product for latency processing in a system. The method involves acquiring a record for a data persistence operation, which includes the actual latency of the operation and measurements of the system's states within a specific time period. If the actual latency exceeds a threshold, an estimated latency is generated using a trained predictor based on the measurements. The difference between the actual and estimated latency is determined, and if it falls below another threshold, one or more states are identified based on the record and estimated latency. This solution helps identify the main factors causing high latency in data persistence operations and provides targeted repair measures to users.

• Method, device, and computer program product for latency processing in a system
• Acquiring a record for a data persistence operation, including actual latency and measurements of system states
• Generating an estimated latency using a trained predictor if the actual latency exceeds a threshold
• Determining the difference between actual and estimated latency
• Identifying one or more states based on the record and estimated latency if the difference is below a threshold

Potential applications of this technology:

• Data persistence systems
• Real-time monitoring and analysis systems
• Performance optimization in computer networks

Problems solved by this technology:

• Identifying factors causing high latency in data persistence operations
• Providing targeted repair measures to address latency issues

Benefits of this technology:

• Improved performance and efficiency in data persistence operations
• Enhanced troubleshooting and problem-solving capabilities
• Cost savings by addressing latency issues in a targeted manner

Abstract

Embodiments of the present disclosure relate to a method, a device, and a computer program product for latency processing. The method includes: in response to that a data persistence operation occurs in a system, acquiring a record for the operation, wherein the record includes an actual latency of the operation and a group of measurements of a group of states of the system within a predetermined time period of the occurrence of the operation. The method further includes: in response to that the actual latency of the operation is greater than a first threshold, generating an estimated latency of the operation by using a trained predictor on the basis of the group of measurements. The method further includes: determining a difference between the actual latency and the estimated latency, and in response to that the difference is less than a second threshold, identifying one or more states from the group of states on the basis of the record and the estimated latency. When there is a high-latency problem in the data persistence operation of the system, the solution of the present disclosure can identify the most major factors that lead to this problem, and thus can provide targeted repair measures to users.

DYNAMIC DEVICE ENABLEMENT (17838423)

Main Inventor

Srikanth Kondapi

Brief explanation

Abstract:

Dynamic device enablement is provided. A device management application can be configured as a shell application and can be customized dynamically with plugins for devices that are connected to a computing device. The plugins can be defined within driver metadata of drivers for the devices. When the drivers are loaded, the device management application can access the driver metadata to dynamically determine which plugins to install and can then automatically install them.

Patent/Innovation Explanation:

• Device management application can be customized with plugins for connected devices.
• Plugins are defined within driver metadata of the devices.
• Device management application accesses driver metadata to determine which plugins to install.
• Automatic installation of the determined plugins.

Potential Applications:

• Device management and enablement in various computing devices.
• Customization of device management applications based on connected devices.

Problems Solved:

• Simplifies the process of managing and enabling devices connected to a computing device.
• Eliminates the need for manual installation of plugins for each device.

Benefits:

• Streamlines device management by automatically installing necessary plugins.
• Reduces the effort required to enable and manage devices.
• Enhances user experience by providing a dynamic and customized device management application.

Abstract

Dynamic device enablement is provided. A device management application can be configured as a shell application and can be customized dynamically with plugins for devices that are connected to a computing device. The plugins can be defined within driver metadata of drivers for the devices. When the drivers are loaded, the device management application can access the driver metadata to dynamically determine which plugins to install and can then automatically install them.

AUTOMATED TESTCASE EXECUTION (17837418)

Main Inventor

Subhashini Puram

Brief explanation

The abstract describes a method for generating a control file and a repository class based on the selection of user interface (UI) controls in an application. 

• The method involves detecting the selection of one or more UI controls in the application's UI.
• It retrieves the respective property value set for each selected UI control, which includes the values of properties associated with the control.
• A control file is generated, containing UI control definitions for each selected control. Each definition is based on the respective property value set of the corresponding control.
• A repository class is generated, which includes handles that can return an object corresponding to each selected UI control.

Potential applications of this technology:

• This method can be used in software development to automate the generation of control files and repository classes for UI controls.
• It can simplify the process of managing and manipulating UI controls in an application.

Problems solved by this technology:

• Manual creation of control files and repository classes for UI controls can be time-consuming and prone to errors. This method automates the generation process, reducing the effort and potential mistakes.

Benefits of this technology:

• Automation of control file and repository class generation saves time and effort in software development.
• The method ensures consistency and accuracy in the generation of control files and repository classes.
• It simplifies the management and manipulation of UI controls in an application.

Abstract

A method, comprising: detecting a selection of one or more user interface (UI) controls in a user interface of an application; retrieving a respective property value set for each of the selected UI controls, the respective property value set for any of the selected UI controls including respective values of one or more properties of the selected UI control; generating a control file, the control file including one or more UI control definitions, each of the UI control definitions corresponding to a different one of the selected UI controls, each of the UI control definitions being generated based on the respective property value set for the UI control that corresponds to the UI control definition; and generating a repository class including one or more handles, each of the handles being configured to return an object corresponding to a different one of the selected UI controls.

QUALITY CONTROL SYSTEM FOR QUANTUM-AS-A-SERVICE BROKERS (17836683)

Main Inventor

Brendan Burns Healy

Brief explanation

The patent application describes quality control operations for quantum processing units. 

• A broker application is used to perform quality control jobs on quantum processing units.
• Baseline characteristics about the quantum processing units are determined through these quality control jobs.
• User-submitted jobs are compared to the baseline characteristics to generate a confidence score.
• The confidence score indicates whether the user-submitted job was performed on the expected quantum processing unit.

Potential Applications

• Ensuring that user-submitted jobs are performed on the correct quantum processing unit.
• Improving the reliability and accuracy of quantum processing units.
• Streamlining quality control operations for quantum processing units.

Problems Solved

• Ensuring that user-submitted jobs are performed on the intended quantum processing unit.
• Identifying any discrepancies between user-submitted jobs and baseline characteristics.
• Enhancing the trustworthiness and credibility of quantum processing units.

Benefits

• Increased confidence in the accuracy and reliability of quantum processing units.
• Improved quality control processes for quantum processing units.
• Enhanced user experience by ensuring that jobs are performed on the expected quantum processing unit.

Abstract

Quality control operations for quantum processing units are disclosed. A broker application may perform quality control jobs to determine baseline characteristics about quantum processing units. The characteristics of user-submitted jobs can be compared to the baseline characteristics associated with the quality control jobs. This allows the broker to generate a confidence score that reflects at least whether the user-submitted job was performed on the quantum processing unit expected by the user.

CREDIT-BASED SCHEDULING USING LOAD PREDICTION (17839943)

Main Inventor

Philippe Armangau

Brief explanation

The abstract describes a technique for scheduling tasks in a storage system by predicting excess capacity of processing resources and determining a credit based on the prediction. This credit is used to prioritize speed-critical tasks over speed-noncritical tasks.

• Predicts excess capacity of processing resources over a known interval of time
• Determines a quantity of credit based on the predicted capacity
• Holds back an increase in priority for speed-noncritical tasks by consuming a portion of the credit
• Allows speed-critical tasks to run with undiminished access to the processing resources

Potential Applications

This technique can be applied in various storage systems where task scheduling is crucial, such as:

• Cloud computing platforms
• Data centers
• Distributed file systems

Problems Solved

The technique addresses the following problems in task scheduling:

• Efficiently utilizing processing resources
• Prioritizing speed-critical tasks
• Preventing speed-noncritical tasks from hindering the performance of speed-critical tasks

Benefits

The benefits of this technique include:

• Improved performance of speed-critical tasks
• Optimal utilization of processing resources
• Enhanced overall system efficiency

Abstract

A technique for scheduling tasks in a storage system includes predicting excess capacity of processing resources over a known interval of time and determining a quantity of credit based on the predicted capacity. The technique further includes holding back a requested increase in the priority of one or more speed-noncritical tasks by consuming a portion of the credit and thus allowing one or more speed-critical tasks to run with undiminished access to the processing resources.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TASK ALLOCATION (17852528)

Main Inventor

Zijia Wang

Brief explanation

The abstract describes a method, electronic device, and computer program for task allocation. Here is a simplified explanation of the abstract:

• The method involves obtaining the initial state of a group of dedicated processing sources.
• In response to a request for a group of tasks, the method determines a set of candidate allocations, which indicate how the tasks can be allocated to the dedicated processing resources.
• Based on the initial state, the method determines the operating state of the dedicated processing resources when executing the tasks according to each candidate allocation.
• The method then selects a target allocation from the set of candidate allocations based on the operating state, enabling the tasks to be executed on the dedicated processing resources according to the target allocation.

Potential applications of this technology:

• Task allocation in distributed computing systems.
• Resource management in cloud computing environments.
• Load balancing in server clusters.

Problems solved by this technology:

• Efficient allocation of tasks to dedicated processing resources.
• Optimizing resource utilization and performance.
• Reducing bottlenecks and improving overall system efficiency.

Benefits of this technology:

• Improved task execution efficiency.
• Enhanced resource management and allocation.
• Increased system performance and scalability.

Abstract

Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for task allocation. The method includes: obtaining an initial state of a group of dedicated processing sources; determining, in response to a request for receiving a group of tasks, a set of candidate allocations each indicating allocation of the group of tasks to the group of dedicated processing resources; determining, based on the initial state, an operating state of the group of dedicated processing resources when executing the group of tasks according to each candidate allocation in the set of candidate allocations; and selecting a target allocation from the set of candidate allocations based on the operating state to enable the group of tasks to be executed on the group of dedicated processing resources according to the target allocation.

ATTRIBUTES FOR WORKLOADS, INFRASTRUCTURE, AND DATA FOR AUTOMATED EDGE DEPLOYMENT (17806857)

Main Inventor

Eric Bruno

Brief explanation

The patent application describes a method for placing workloads in a computing environment based on their attributes and the attributes of the nodes and data. Here are the key points:

• Workloads in a computing environment are given a workload score based on their attributes.
• The workload attributes are compared with the attributes of the nodes and/or data.
• Based on this comparison and the node score, the workload is placed with one of the nodes.
• This attribute-based workload placement helps optimize resource allocation in the computing environment.

Potential applications of this technology:

• Cloud computing platforms can use this method to efficiently allocate workloads to different nodes based on their attributes.
• Data centers can benefit from this technology by optimizing workload placement and improving overall performance.
• Distributed computing systems can use this method to balance the workload across different nodes based on their capabilities.

Problems solved by this technology:

• Efficient workload placement: The method helps in placing workloads in the most suitable nodes based on their attributes, leading to better resource utilization.
• Resource optimization: By considering both workload and node attributes, the method ensures that workloads are placed in nodes that can handle them effectively, reducing resource wastage.
• Scalability: The attribute-based approach allows for easy scalability as new nodes can be added to the computing environment and workloads can be dynamically placed based on their attributes.

Benefits of this technology:

• Improved performance: By placing workloads in nodes that are best suited for them, overall system performance is enhanced.
• Resource efficiency: The attribute-based placement ensures that resources are utilized optimally, reducing costs and improving energy efficiency.
• Flexibility and scalability: The method allows for dynamic workload placement, making it adaptable to changing computing environments and workload demands.

Abstract

Attribute-based workload placement and orchestration in a computing environment including nodes is disclosed. A workload, when received at a scheduling engine, is given a workload score determined from the workload's attributes. Using the workload attributes, along with node attributes and/or data attributes, the workload is placed with one of the nodes. The node is selected based on how the workload attributes compare with the node attributes and/or the data attributes and based on the node score.

LCS WORKLOAD IN-BAND RESOURCE DEVICE MANAGEMENT SYSTEM (17836329)

Main Inventor

Srinivas Giri Raju Gowda

Brief explanation

The abstract describes a Logically Composed System (LCS) workload in-band resource device management system. It includes a resource device management subsystem, a workload using LCS, and a System Control Processor (SCP) device. The resource device management subsystem identifies available resource devices through the SCP device and presents them to the workload. When the workload requests a resource device, the subsystem sends a provisioning request to the SCP device to provide the requested resource device to the LCS.

• The system manages resource devices in an LCS workload.
• It includes a resource device management subsystem, workload, and SCP device.
• The subsystem identifies available resource devices and presents them to the workload.
• When a resource device is requested, a provisioning request is sent to the SCP device.

Potential Applications

• Cloud computing platforms
• Virtualization environments
• Distributed computing systems

Problems Solved

• Efficient management of resource devices in an LCS workload
• Streamlining resource provisioning process
• Simplifying resource device allocation and utilization

Benefits

• Improved resource management and utilization
• Enhanced scalability and flexibility in workload provisioning
• Reduced complexity in managing resource devices

Abstract

A Logically Composed System (LCS) workload in-band resource device management system includes a resource device management subsystem coupled to a workload that is provided using an LCS, and to a System Control Processor (SCP) device. The resource device management subsystem determines a first plurality of resource devices that are available via the SCP device for the LCS, and presents an available-resource file hierarchy to the workload that identifies the first plurality of resource devices. When the resource device management subsystem receives a resource device request from the workload for a first resource device that is included in the first plurality of resource devices identified in the available-resource file hierarchy, it provides a resource device provisioning request to the SCP device that is configured to cause the SCP device to provide the first resource device to the LCS for utilization by the workload.

LCS WORKLOAD IN-BAND SERVICE MANAGEMENT SYSTEM (17837285)

Main Inventor

Srinivas Giri Raju Gowda

Brief explanation

The abstract describes a Logically Composed System (LCS) workload in-band service management system. It consists of a service management subsystem, a workload using LCS, and a System Control Processor (SCP) device. The service management subsystem identifies and presents a service library to the workload, which includes a variety of services provided by or available via the SCP device. When the workload requests a service from the library, the service management subsystem sends a provisioning request to the SCP device to provide the requested service to the LCS for utilization by the workload.

• The system manages services provided by or available via the SCP device for use by the LCS workload.
• The service management subsystem presents a service library to the workload, identifying the available services.
• When the workload requests a service, the service management subsystem sends a provisioning request to the SCP device.
• The SCP device provides the requested service to the LCS for utilization by the workload.

Potential Applications

• This technology can be applied in cloud computing environments to manage and provision services for workloads.
• It can be used in software-defined networking (SDN) to manage and provision network services for different applications.
• The system can be utilized in virtualized environments to efficiently allocate and manage resources for workloads.

Problems Solved

• Simplifies the management and provisioning of services for workloads in an LCS environment.
• Provides a centralized system for identifying, presenting, and provisioning services.
• Streamlines the process of requesting and utilizing services by the workload.

Benefits

• Increases efficiency by automating the provisioning of services for workloads.
• Reduces manual effort and potential errors in managing and provisioning services.
• Enables better resource allocation and utilization in an LCS environment.

Abstract

A Logically Composed System (LCS) workload in-band service management system includes a service management subsystem coupled to a workload that is provided using a Logically Composed System (LCS), and a System Control Processor (SCP) device. The service management subsystem determines a plurality of services that are either provided by or available via the SCP device for use by the LCS, and presents a service library to the workload that identifies the plurality of services. If the service management subsystem receives a first service request from the workload for a first service that is included in the plurality of service identified in the service library, it provides a first service provisioning request to the SCP device that is configured to cause the SCP device to provide the first service to the LCS for utilization by the workload.

SELF-HEALING RECOVERY OF FILES USING A CYBER RECOVERY VAULT FOR DEDUPLICATION FILE SYSTEMS (17839192)

Main Inventor

Nitin Madan

Brief explanation

The abstract describes a self-healing method for recovering files using a cyber recovery vault for deduplication filesystems. Here are the key points:

• Data from a production system is backed up to a backup system in a datacenter and also copied to a cyber security vault.
• The copy process creates a point-in-time (PIT) copy of the data in the vault and locks it for retention.
• An automated process detects missing segment errors, such as bad index lookups in the data being backed up.
• The self-healing process retrieves the missing segment from the PIT copy in the cyber security vault.
• For any missing fingerprints, the process adds the data in new containers and updates the index.
• This allows for efficient recovery by writing new containers with the missing references and updating the index.

Potential applications of this technology:

• Data backup and recovery systems in various industries, such as healthcare, finance, and government.
• Cybersecurity solutions for protecting critical data and ensuring its availability in case of errors or attacks.

Problems solved by this technology:

• Efficient recovery of missing data segments in deduplication filesystems.
• Ensuring data integrity and availability in backup systems.
• Streamlining the backup and recovery process by automating error detection and self-healing.

Benefits of this technology:

• Improved data protection and recovery capabilities.
• Reduced downtime and data loss in case of errors or attacks.
• Simplified management of backup systems through automated processes.

Abstract

A self-healing method of recovering files using a cyber recovery vault for deduplication filesystems. Data to be backed up from a production system to a backup system in a datacenter is also copied to a cyber security vault coupled to the data center through an air gapped connection. A copy process makes a point-in-time (PIT) copy of the portion of data to a namespace in the vault, and retention locks the copied portion in the vault. An automated process detects a missing segment error, such as due to a bad index lookup in the data to be backed up. The self-healing process provides the missing segment from the PIT copy in the cyber security vault. For any missing fingerprints the process adds the data in new containers and the references in the index, providing an indirection that allows for just writing a new container with the missing L0 references and updating the index.

METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR PROCESSING SYSTEM LATENCY (17875202)

Main Inventor

Bing LIU

Brief explanation

The present disclosure describes a method, device, and computer program product for processing system latency in a complex system environment. The method involves obtaining a group of records related to data persistence operations, estimating latencies based on these records, and determining the contributions of different states in the system to these latencies. The method also identifies the major factors that increase latencies and their degree of influence on latency. This information can help users in improving the system's performance by targeting specific areas.

• The method involves obtaining a group of records related to data persistence operations.
• Latencies of the data persistence operations are estimated based on these records.
• The contributions of different states in the system to the latencies are determined.
• Major factors that increase latencies and their degree of influence on latency are identified.
• The method helps users in identifying areas for targeted improvement in a complex system environment.

Potential Applications

This technology can be applied in various fields where system latency is a concern, such as:

• Cloud computing
• Database management systems
• Network infrastructure
• Real-time data processing systems

Problems Solved

The technology solves the following problems:

• Identifying major factors that increase latencies in a complex system environment.
• Determining the degree of influence of these factors on latency.
• Providing users with targeted improvement directions to reduce latencies in data persistence operations.

Benefits

The benefits of this technology include:

• Improved system performance by reducing latencies in data persistence operations.
• Targeted improvement directions based on the major factors contributing to latency.
• Enhanced efficiency and responsiveness of complex systems.

Abstract

Embodiments of the present disclosure relate to a method, a device, and a computer program product for processing a system latency. The method includes obtaining a group of records in a system for a group of data persistence operations of a particular type, and estimating a group of estimated latencies in the group of data persistence operations based on the group of records. Each record in the group of records includes a group of metrics for a group of states of the system within a predetermined period when each data persistence operation occurs. The method also includes determining corresponding contributions of each state in the group of states to latencies of the group of data persistence operations based on the group of records and the group of estimated latencies, and determining one or more states from the group of states based on the corresponding contributions. Embodiments of the present disclosure can identify, in a complex system environment, major factors that increase the latencies of data persistence operations and the degree of influence on the amount of latency by these factors, thereby providing a user with a targeted improvement direction.

EFFICIENT METHOD TO DYNAMICALLY SELECT A PROTECTION DURATION FOR RETENTION LOCKING DEDUPLICATED OBJECTS (17805908)

Main Inventor

Jagannathdas Rath

Brief explanation

==Abstract==

The patent application describes a method for dynamically selecting the duration of locks on objects stored in a storage system. This is done in order to optimize costs and ensure that objects are not removed from storage until their associated locks have expired. The method takes into account factors such as garbage collection operations, the number of deleted objects that are still stored due to being locked, and the churn rate of the garbage collection process. By dynamically and repeatedly selecting the most suitable lock extension value, the system can automatically control costs.

Explanation

• The patent application proposes a method for selecting the duration of locks on objects in a storage system.
• The purpose is to optimize costs and prevent objects from being removed from storage before their locks expire.
• The method considers factors such as garbage collection operations, the number of locked objects still stored, and the garbage collection churn rate.
• By dynamically selecting the most appropriate lock extension value, costs can be controlled automatically.

Potential Applications

• This technology can be applied in any storage system where objects need to be locked and protected for a certain duration.
• It can be particularly useful in cloud storage systems, databases, and distributed file systems.
• The method can also be implemented in backup and recovery systems to ensure locked objects are not prematurely deleted.

Problems Solved

• The method solves the problem of objects being removed from storage before their associated locks have expired.
• It addresses the challenge of optimizing costs associated with locking objects in a storage system.
• The method also tackles the issue of dynamically adjusting lock durations based on various factors, such as garbage collection operations and churn rate.

Benefits

• The technology provides an automated and dynamic approach to selecting lock durations, reducing the need for manual intervention.
• By optimizing lock durations, costs associated with storage can be minimized.
• The method ensures that locked objects are not deleted prematurely, improving data integrity and reliability.

Abstract

Dynamically selecting a protection duration for locking objects in a storage. A garbage collection deletes objects over time. The objects may not be removed from storage until associated locks have expired. The length of the lock may impact the cost. A new lock extension value is selected each time a lock extension operation is performed. The lock extension value selection process accounts for a garbage collection operation cycle, an amount of deleted objects that are stored because of being locked, and a garbage collection churn. This allows an optimum lock extension value to be selected dynamically and repeatedly to control costs automatically.

DYNAMIC FILESYSTEM GENERATION BASED ON CONTENT METADATA (17806770)

Main Inventor

Joel Christner

Brief explanation

The abstract describes a method in a computer system where an administrator receives a query from a user computer system, parses the query to identify metadata, identifies content in a datastore that matches the metadata, packages the content into a virtual filesystem, and exposes the virtual filesystem to the user computer system.

• The method involves a networked system of user computer systems and an administrator.
• The administrator receives a query from a user computer system, indicating the desired content.
• The query is parsed by the administrator to extract metadata, which provides additional information about the content.
• The administrator searches a datastore for content that matches the metadata.
• The identified content is then packaged into a virtual filesystem, which is a simulated file system that can be accessed by the user computer system.
• The virtual filesystem is exposed to the user computer system, allowing the user to access and interact with the content.

Potential Applications

• Content management systems: This method can be used to efficiently organize and provide access to large amounts of content in a networked system.
• Data retrieval systems: It can be applied in systems where users need to search and retrieve specific data based on metadata.
• Collaborative platforms: The method can facilitate sharing and accessing content among multiple users in a collaborative environment.

Problems Solved

• Efficient content retrieval: The method streamlines the process of searching and retrieving content by utilizing metadata and packaging it into a virtual filesystem.
• Scalability: The system can handle a large number of user computer systems and efficiently manage and expose the content to them.
• Simplified access: Users can easily access and interact with the content through the virtual filesystem, eliminating the need for complex retrieval processes.

Benefits

• Improved user experience: Users can quickly and easily access the desired content through a familiar file system interface.
• Enhanced content organization: The method leverages metadata to categorize and organize content, making it easier to locate and manage.
• Efficient resource utilization: By packaging content into a virtual filesystem, the system optimizes resource allocation and reduces the need for redundant storage.

Abstract

One example method includes receiving, by an administrator from a user computer system, a query identifying content the user wants to access, and the administrator and user computer system are elements of a system that includes a group of networked user computer systems, parsing, by the administrator, the query to identify metadata included in the query, identifying, by the administrator, any content of a datastore that matches the metadata, packaging, by the administrator, the datastore content into a virtual filesystem, and exposing, by the administrator, the virtual filesystem to one of the user computer systems.

EXTENDING METADATA-DRIVEN CAPABILITIES IN A METADATA-CENTRIC FILESYSTEM (17806773)

Main Inventor

Joel Christner

Brief explanation

The abstract of this patent application describes a method for managing data by declaring a requirement for the data, defining metadata that represents the requirement, associating the requirement with the data, and performing filesystem operations based on the metadata. The method is performed by an administrator using a centralized datastore with a metadata repository.

• The method involves declaring a management requirement for a collection of data.
• Metadata is defined to represent the requirement.
• The requirement is associated with the data.
• Filesystem operations are performed on the data based on the metadata.
• An administrator operates in a centralized datastore with a metadata repository.
• The defined metadata is stored in the metadata repository.

Potential Applications

• Data management in large organizations or enterprises.
• Filesystem operations automation based on metadata.
• Centralized data management systems.

Problems Solved

• Efficient management of data based on specific requirements.
• Streamlining filesystem operations through metadata association.
• Centralized administration and control of data.

Benefits

• Improved organization and management of data.
• Automation of filesystem operations based on metadata.
• Enhanced efficiency and productivity in data management.
• Centralized control and administration of data.

Abstract

One example method includes declaring a management requirement for a collection of data, defining metadata that embodies the requirement, associating the management requirement with the collection of data, and performing a filesystem operation, with respect to the collection of data, based on the metadata. The declaring and defining operations may be performed by an administrator operating in a centralized datastore that includes a metadata repository, and the defined metadata may be stored in the metadata repository.

AN EFFICIENT METHOD TO FIND COLUMNS OF A TABLE WHICH ARE UNIQUE (17806549)

Main Inventor

Seth Jacob Rothschild

Brief explanation

The abstract of this patent application describes a method for determining whether a column in a table can be considered as a primary key. Here is a simplified explanation:

• The method starts by identifying a column in a table that contains multiple entries.
• A proposed uniqueness is set for the column, which means determining how unique the values in the column should be.
• A confidence tolerance is set for the proposed uniqueness, which determines the level of certainty required for considering the column as a primary key.
• Based on the proposed uniqueness and confidence tolerance, an estimated sub-sample size is calculated for the column. This determines the number of entries that need to be sampled.
• Using the sub-sample size, a subset of entries is randomly sampled from the column.
• Based on the sampled subset, it is determined whether the column can be considered as a primary key for the table.

Potential applications of this technology:

• Database management systems can use this method to automatically determine primary keys for tables, saving time and effort for database administrators.
• Data analysis and data mining applications can benefit from this method to identify unique identifiers in large datasets.

Problems solved by this technology:

• Manual identification of primary keys in large tables can be time-consuming and prone to errors. This method automates the process and provides a reliable way to determine primary keys.
• Ensuring the uniqueness of values in a column is crucial for maintaining data integrity. This method helps in verifying the uniqueness of a column and identifying potential primary keys.

Benefits of this technology:

• Saves time and effort by automating the process of determining primary keys.
• Provides a reliable and systematic approach to ensure the uniqueness of values in a column.
• Reduces the risk of errors in identifying primary keys, leading to improved data integrity.

Abstract

One example method includes identifying a column of a table, and the column includes multiple entries, setting a proposed uniqueness for the column, setting a confidence tolerance for the proposed uniqueness, estimating a sub-sample size for the column based on the proposed uniqueness and the confidence tolerance, based on the sub-sample size, sampling a subset of the entries in the column and, based on the sampling, determining whether or not the column is a primary key for the table.

NATIVE METADATA GENERATION WITHIN A STREAM-ORIENTED SYSTEM (17806767)

Main Inventor

Joel Christner

Brief explanation

The abstract of this patent application describes a method performed by a computer system in a network. The method involves intercepting stream elements, selecting and analyzing them, and then emitting them to an external storage entity. The emitted data may include both the selected stream elements and metadata generated by the computer system.

• The computer system intercepts stream elements from a stream associated with it.
• It selects specific stream elements for further analysis.
• The selected stream elements are analyzed by the computer system.
• Based on the analysis, the computer system emits the selected stream elements to an external storage entity.
• The emitted data may include both the selected stream elements and metadata generated by the computer system.

Potential Applications

• Data analysis and storage in real-time streaming applications.
• Monitoring and analyzing network traffic for security purposes.
• Capturing and analyzing user behavior on a website or application.

Problems Solved

• Efficiently capturing and analyzing stream elements in a network environment.
• Storing and querying large amounts of data generated by a computer system.
• Integrating natively generated metadata with the selected stream elements.

Benefits

• Real-time analysis and storage of streaming data.
• Improved network security through monitoring and analysis.
• Enhanced understanding of user behavior and preferences.
• Efficient storage and retrieval of large amounts of data.

Abstract

One example method includes, at a computer system that is an element of a network, intercepting stream elements of a stream associated with the computer system, selecting one or more of the stream elements, analyzing the selected stream elements, and based on the analyzing, emitting the selected stream elements to an external, query-able, storage entity that is external to the computer system. The metadata emitted to the storage entity may further include metadata that was natively generated at the computer system.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR DATA PROCESSING (17854383)

Main Inventor

Si Chen

Brief explanation

The abstract describes a method, electronic device, and computer program for data processing. It involves receiving a data conversion strategy from a server, determining metadata of unstructured data based on the strategy, and transmitting the metadata to the server. This enables edge computing on unstructured data, leading to timely processing and reduced computing load on the server side, improving user experience.

• The method involves receiving a data conversion strategy from a server.
• Unstructured data from a field device is received and metadata is determined based on the data conversion strategy.
• A set of metadata is formed and at least a part of it is transmitted to the server.
• Edge computing is performed on unstructured data, allowing for timely processing of monitoring data.
• Computing load on the server side is reduced, leading to an improved user experience.

Potential Applications

• This technology can be applied in various industries where real-time data processing is required, such as manufacturing, healthcare, and transportation.
• It can be used in IoT (Internet of Things) systems to process data collected from sensors and devices.
• The method can be implemented in data analytics platforms to efficiently handle large volumes of unstructured data.

Problems Solved

• The technology solves the problem of timely processing of unstructured data by enabling edge computing, reducing the reliance on server-side processing.
• It addresses the issue of high computing load on the server side by performing data processing at the edge, improving the overall user experience.

Benefits

• Timely processing of monitoring data allows for quick decision-making and response to events.
• Reduced computing load on the server side leads to improved system performance and scalability.
• Edge computing enables real-time data processing, enhancing the efficiency and effectiveness of data analytics applications.

Abstract

Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for data processing. The method may include receiving a data conversion strategy from a server. The method may further include determining, in response to receiving unstructured data from a field device, metadata of the unstructured data based on the received data conversion strategy to form a set of metadata. In addition, the method may include transmitting at least a part of the set of metadata to the server. According to embodiments of the present disclosure, edge computing can be performed on unstructured data, which not only enables timely processing of monitoring data, but also reduces computing load on the side of the server, thereby improving the user experience.

METHOD, SERVER, AND SYSTEM FOR IMAGE DEPLOYMENT (17860764)

Main Inventor

Anzhou Hou

Brief explanation

The abstract describes a method for efficiently pulling images from multiple nodes in a repository server. Here is a simplified explanation of the abstract:

• The repository server receives requests from multiple nodes to pull images.
• It detects that the requested images have a common layer among their layers.
• The repository server generates a transmission path for the common layer, where it sends the data of the common layer to a second node.
• The second node then distributes the data of the common layer to other nodes.
• The repository server sends setting information to the first node, indicating how to acquire the data of the common layer.
• The repository server sends the data of the common layer to the second node.
• The first node acquires the data of the common layer from a third node based on the setting information.

Potential applications of this technology:

• Cloud computing: Efficiently pulling images from multiple nodes in a cloud environment can improve resource management and reduce network traffic.
• Content delivery networks: Pulling common layers of images from a central repository can optimize content distribution across multiple nodes in a CDN.
• Software updates: Pulling common layers of software images from a repository can speed up the update process for multiple devices.

Problems solved by this technology:

• Reducing network traffic: By identifying and distributing common layers of images, the method minimizes the amount of data that needs to be transferred over the network.
• Resource optimization: The method allows for efficient use of resources by avoiding redundant transfers of common layers.
• Faster image retrieval: By providing a transmission path and setting information, the method enables nodes to acquire common layers more quickly.

Benefits of this technology:

• Improved efficiency: The method optimizes the pulling process by identifying and distributing common layers, reducing network congestion and resource usage.
• Faster image retrieval: By providing a transmission path and setting information, the method enables nodes to retrieve common layers more efficiently, resulting in faster overall image pulling.
• Cost savings: By minimizing network traffic and resource usage, the method can lead to cost savings in terms of bandwidth and infrastructure requirements.

Abstract

A method includes: receiving, by a repository server, requests for pulling images from a plurality of nodes, wherein an image requested to be pulled by a first node is a first image; detecting, by the repository server, that layers of a plurality of images requested to be pulled by the plurality of nodes all include a common layer; generating, by the repository server, a transmission path of data of the common layer, wherein the transmission path includes that the repository server transmits the data of the common layer to a second node, and the second node distributes the data of the common layer to other nodes; sending, by the repository server, setting information to the first node according to the transmission path; sending, by the repository server, the data of the common layer to the second node; and acquiring, by the first node, the data of the common layer from a third node according to the setting information.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR DETECTING MODEL DRIFT (17857222)

Main Inventor

Jiacheng Ni

Brief explanation

The patent application describes a method for detecting model drift in a decision tree model used for image classification, text classification, or data mining. The method involves converting training data into an input vector represented by Shapley values, where each dimension of the vector represents an input feature of the model. The training data is then clustered based on the input vector to obtain data clusters. 

• The method converts training data into an input vector represented by Shapley values.
• The input vector represents the input features of a decision tree model used for image classification, text classification, or data mining.
• The training data is clustered based on the input vector to obtain data clusters.
• The method can detect the drift degree of the decision tree model by comparing a new input vector with the existing data clusters.

Potential applications of this technology:

• Monitoring and detecting model drift in image classification systems.
• Monitoring and detecting model drift in text classification systems.
• Monitoring and detecting model drift in data mining systems.

Problems solved by this technology:

• Model drift can occur over time due to changes in the input data, leading to degraded performance of the decision tree model.
• Detecting model drift manually can be time-consuming and subjective.
• This technology provides an automated method to detect model drift and assess its degree.

Benefits of this technology:

• Improved accuracy and performance of decision tree models by detecting and addressing model drift.
• Time and cost savings by automating the detection of model drift.
• Objective and consistent assessment of model drift degree.

Abstract

In a method for detecting a model drift provided in an illustrative embodiment of the present disclosure, training data in a training data set is converted into an input vector represented by Shapley values. A plurality of dimensions of the input vector indicates a plurality of input features of a decision tree model. The decision tree model has been trained for performing at least one of image classification, text classification, or data mining. The method also includes: clustering, on the basis of the input vector, the training data set, so as to obtain a plurality of data clusters. The method also includes: in response to receiving a first input, converting the first input into a first input vector represented by Shapley values. The method also includes: detecting a drift degree of the decision tree model on the basis of the first input vector and the plurality of data clusters.

PRE-AUTHORIZED VIRTUALIZATION ENGINE FOR DYNAMIC FIRMWARE MEASUREMENT (17837329)

Main Inventor

Shekar Babu SURYANARAYANA

Brief explanation

The abstract describes a virtual BIOS engine that can emulate firmware update processes using virtual computer-readable media and a virtual cryptoprocessor. It can extract firmware payload and measure it using a virtual trust chain.

• The virtual BIOS engine is configured to emulate firmware update processes during the runtime of an operating system.
• It can load onto an isolated compute domain of the processor to perform the emulation.
• The emulation includes using virtual computer-readable media and a virtual cryptoprocessor.
• The firmware payload is extracted and stored in the virtual computer-readable media.
• A virtual trust chain is executed to measure the firmware payload.

Potential Applications

• This technology can be used in computer systems to securely update firmware.
• It can be applied in various industries such as IT, telecommunications, and automotive.

Problems Solved

• The virtual BIOS engine solves the problem of securely updating firmware without the need for physical media or hardware.
• It provides a way to emulate firmware update processes in a controlled and isolated environment.

Benefits

• The use of virtual computer-readable media and a virtual cryptoprocessor enhances security and reduces the risk of firmware tampering.
• The virtual BIOS engine allows for efficient and streamlined firmware updates.
• It provides a flexible and scalable solution for firmware management.

Abstract

A virtual BIOS engine may be configured to, during runtime of an operating system, in response to an operating system event for updating firmware, load onto an isolated compute domain of the processor to emulate firmware update processes of a non-transitory computer-readable media with a virtual non-transitory computer-readable media and emulate the firmware update processes of the cryptoprocessor with a virtual cryptoprocessor, extract a firmware payload to the virtual non-transitory computer-readable media, and execute a virtual trust chain to measure the firmware payload in the virtual non-transitory computer-readable media.

METHODS AND SYSTEMS FOR GENERATING FORECASTS USING AN ENSEMBLE ONLINE DEMAND GENERATION FORECASTER (17746710)

Main Inventor

Arun Kumar Venkitaraman

Brief explanation

The abstract describes a method for generating final prediction data using a prediction model. Here are the key points:

• The method involves generating validation prediction data using a prediction model.
• It identifies unexplained variance data within the validation prediction data.
• Initial prediction data is generated using the prediction model.
• Prediction trend data is isolated from the initial prediction data.
• The final prediction data is obtained by summing the prediction trend data with the unexplained variance data.
• A user can provide a constraint for the final prediction data.
• The method determines if the final prediction data satisfies the constraint.
• If the constraint is satisfied, the method indicates to the user that the final prediction data meets the constraint.

Potential Applications

• This method can be applied in various fields that require prediction models, such as finance, weather forecasting, and sales forecasting.
• It can be used in data analysis and decision-making processes where accurate predictions are crucial.

Problems Solved

• The method addresses the issue of unexplained variance in prediction data, which can affect the accuracy and reliability of predictions.
• By isolating and incorporating the unexplained variance data into the final prediction, the method improves the overall accuracy of the predictions.

Benefits

• The method provides a more accurate and reliable prediction by considering both the prediction trend and the unexplained variance.
• Users can set constraints to ensure that the final prediction data meets specific requirements.
• It allows for better decision-making based on more accurate predictions.

Abstract

A method for generating final prediction data, that includes generating, by a prediction generator, validation prediction data using a prediction model, making a first determination that the validation prediction data includes unexplained variance data, in response to making the first determination, isolating unexplained variance data from the validation prediction data, generating initial prediction data using the prediction model, isolating prediction trend data from the initial prediction data, obtaining the final prediction data by summing the prediction trend data with the unexplained variance data, receiving, from a user, a constraint, making a second determination that the final prediction data satisfies the constraint, and based on the second determination, indicating, to the user, that the final prediction data satisfies the constraint.

METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TRAINING IMAGE CLASSIFICATION MODEL (17850545)

Main Inventor

Bin He

Brief explanation

The abstract describes a method, device, and computer program for training an image classification model. Here are the key points:

• The method involves training an image classification model to classify pixel points of a sample image into different object classes.
• In each training iteration, the model determines a classification result for the pixel points.
• The first classification result is obtained in the first training iteration, and the second classification result is obtained in the second training iteration.
• Based on these classification results, the method determines recall rates for each object class in both iterations.
• The recall rates represent the model's ability to correctly identify instances of each object class.
• The image classification model is then adjusted based on these recall rates to improve its performance.
• The result is a trained image classification model that can accurately classify objects in images.

Potential applications of this technology:

• Image recognition and classification systems
• Object detection in images and videos
• Autonomous vehicles for identifying and tracking objects on the road
• Surveillance systems for detecting specific objects or individuals

Problems solved by this technology:

• Improves the accuracy and performance of image classification models
• Enables more precise identification and classification of objects in images
• Reduces false positives and false negatives in object detection systems

Benefits of this technology:

• Enhanced accuracy in identifying and classifying objects in images
• Improved performance of image classification models
• Increased reliability and efficiency in object detection systems
• Enables the development of more advanced and reliable computer vision applications.

Abstract

Embodiments disclosed herein relate to a method, a device, and a computer program product for training an image classification model. The method includes: determining a first classification result obtained by the image classification model on pixel points of a sample image in a first training iteration, wherein the first classification result indicates that each of the pixel points belongs to one of a plurality of object classes; determining a second classification result obtained by the image classification model on the pixel points of the sample image in a second training iteration; determining, based on the first classification result and the second classification result, a first set of recall rates and a second set of recall rates for the plurality of object classes; and adjusting, based on the first set of recall rates and the second set of recall rates, the image classification model to obtain a trained image classification model.

QUESTION ANSWERING METHOD, ELECTRONIC DEVICE, AND PROGRAM PRODUCT (17866702)

Main Inventor

Zijia Wang

Brief explanation

The present disclosure describes a method, electronic device, and computer program product for question answering. The method involves determining an answer library associated with a question, determining a restricted Boltzmann machine associated with the answer library, and determining annotation information using the restricted Boltzmann machine.

• The method determines an answer library associated with a question.
• The method determines a restricted Boltzmann machine associated with the answer library.
• The restricted Boltzmann machine determines a set of questions that the answer library can answer.
• The restricted Boltzmann machine also determines association relationships between questions in the set and the answer library.
• The method determines annotation information associated with the question and targeted to the answer library using the restricted Boltzmann machine.

Potential applications of this technology:

• Question answering systems in customer service or technical support.
• Chatbots or virtual assistants that can provide accurate and targeted answers.
• Information retrieval systems that can quickly find relevant answers to user queries.

Problems solved by this technology:

• Improves the accuracy and efficiency of question answering systems.
• Enables customer service personnel to have a better understanding of the answer library.
• Provides targeted recommendation information to enhance user experience.

Benefits of this technology:

• Increases the effectiveness of customer service interactions.
• Saves time and effort by providing accurate and relevant answers.
• Enhances user satisfaction by delivering personalized recommendations.

Abstract

Embodiments of the present disclosure relate to a question answering method, an electronic device, and a computer program product. The method includes: determining an answer library associated with a question; determining a restricted Boltzmann machine associated with the answer library, wherein the restricted Boltzmann machine is configured to determine a set of questions that the answer library can answer and association relationships between questions in the set of questions and the answer library; and determining, using the restricted Boltzmann machine, annotation information associated with the question and targeted to the answer library. With the technical solution of the present disclosure, it is possible to determine, while determining an answer library associated with the question, annotation information that is targeted to the answer library, and to enable customer service personnel to have a more thorough understanding of the determined answer library and obtain targeted recommendation information.

TRANS-INDUCTOR VOLTAGE REGULATOR USING A NONLINEAR COMPENSATION INDUCTOR (17836484)

Main Inventor

Shiguo Luo

Brief explanation

The abstract describes a trans-inductor voltage regulator (TLVR) that includes multiple voltage converter phases, coupled inductors, and a compensation inductor. The TLVR receives a common input voltage and provides an output voltage. Each coupled inductor consists of a primary winding and a secondary winding, with the primary winding connected to the output of the voltage converter and the load. The compensation inductor is connected to a ground plane and the secondary windings are connected in series with the compensation inductor.

• TLVR includes multiple voltage converter phases, coupled inductors, and a compensation inductor.
• Each coupled inductor has a primary winding and a secondary winding.
• Primary winding of each coupled inductor is connected to the output of the voltage converter and the load.
• Compensation inductor is connected to a ground plane.
• Secondary windings are connected in series with the compensation inductor.

Potential applications of this technology:

• Power supply systems
• Voltage regulation in electronic devices
• Renewable energy systems

Problems solved by this technology:

• Efficient voltage regulation
• Improved power conversion
• Reduction of voltage fluctuations

Benefits of this technology:

• Enhanced power efficiency
• Better voltage stability
• Compact design

Abstract

A trans-inductor voltage regulator (TLVR) includes at least two voltage converter phases configured to receive a common input voltage and to provide an output voltage on an output of the voltage converter, an associated coupled inductor, and a compensation inductor. Each coupled inductor includes a primary winding magnetically coupled to a secondary winding. For each primary winding, a first terminal is coupled to the output of the associated voltage converter, and a second terminal of the primary winding is coupled to a load. A first terminal of the compensation inductor is coupled to a ground plane, and each secondary winding is coupled in series with the compensation inductor, with a last secondary being coupled to the ground plane. The compensation inductor is a nonlinear inductor exhibiting a first inductance level at a first current level, and a second inductance level different from the first inductance level at a second current level different from the first current level.

INTELLIGENT VIRTUAL EVENT ASSISTANT (17806137)

Main Inventor

Bijan Kumar Mohanty

Brief explanation

The disclosed patent application describes a methodology for a virtual event assistant that can join an online meeting and receive real-time content, including audio and video streams. The virtual assistant then generates a summarized version of the meeting based on a transcript, using artificial intelligence techniques. It also determines contextual metadata of the meeting by analyzing the content. The summarized content with contextual metadata can be provided to a user for playback.

• Virtual event assistant joins online meeting and receives real-time content
• Content includes audio and video streams of the meeting
• Virtual assistant generates a summarized version of the meeting based on a transcript
• Artificial intelligence techniques are used to generate the summarized content
• Virtual assistant determines contextual metadata of the meeting by analyzing the content
• Summarized content with contextual metadata can be provided to a user for playback

Potential Applications

• Virtual event assistants can help users quickly review and understand the content of online meetings
• Summarized content with contextual metadata can be used for efficient playback and reference purposes
• This technology can be applied in various industries where online meetings are common, such as business, education, and telemedicine

Problems Solved

• Time-consuming process of reviewing and understanding the content of online meetings
• Difficulty in finding specific information within lengthy meeting recordings
• Lack of contextual information about the meeting content

Benefits

• Saves time by providing a summarized version of the meeting
• Enables efficient playback and reference of meeting content
• Provides contextual metadata for better understanding and organization of meeting content

Abstract

In one aspect, an example methodology implementing the disclosed techniques includes, by a virtual event assistant, joining an online meeting and receiving a content of the online meeting in real-time, the content including an audio stream and a video stream of the online meeting. The method also includes, by the virtual event assistant, generating a summarized content of the online meeting based on a transcript of the online meeting, wherein generating the summarized content includes applying one or more artificial intelligence (AI)-based techniques to the transcript of the online meeting. The method further includes, by the virtual meeting assistant, determining contextual metadata of the online meeting based on analysis of the content of the online meeting. The method may also include providing the summarized content with the contextual metadata of the online meeting for playback by a user, for example.

INTELLIGENT INFRASTRUCTURE MANAGEMENT IN A CLOUD RADIO ACCESS NETWORK (17838624)

Main Inventor

Parminder Singh Sethi

Brief explanation

The patent application describes techniques for intelligent infrastructure management in a radio access network. It involves using machine learning models to analyze operational statistics of baseband units and predict future technical issues. Here is a simplified explanation of the patent application:

• The method obtains operational statistics of baseband units in a radio access network.
• A forecasted data set is generated using a machine learning model to predict future operational statistics of the baseband units.
• The forecasted data set is analyzed using another machine learning model to predict future technical issues.
• Based on the analysis, at least one corrective action is determined for the predicted future occurrence of the technical issue.
• The method then initiates the corrective action.

Potential Applications

This technology can be applied in various industries and scenarios where intelligent infrastructure management is crucial. Some potential applications include:

• Telecommunications: Optimizing the performance and maintenance of radio access networks.
• Internet of Things (IoT): Managing and monitoring large-scale IoT deployments efficiently.
• Smart Cities: Enhancing the management of infrastructure in urban environments.
• Industrial Automation: Improving the maintenance and reliability of industrial networks.

Problems Solved

The techniques described in the patent application address several problems in infrastructure management:

• Proactive Issue Detection: By analyzing operational statistics and predicting future technical issues, potential problems can be identified before they occur.
• Efficient Maintenance: The use of machine learning models enables the determination of corrective actions, allowing for timely and targeted maintenance.
• Enhanced Network Performance: By predicting and addressing technical issues, the overall performance and reliability of the network can be improved.

Benefits

The use of intelligent infrastructure management techniques outlined in the patent application offers several benefits:

• Cost Savings: Proactive issue detection and targeted maintenance can help reduce downtime and minimize costly repairs.
• Improved Efficiency: By automating the analysis and prediction process, resources can be allocated more efficiently, leading to optimized network performance.
• Enhanced Reliability: Predicting and addressing technical issues in advance improves the overall reliability and availability of the infrastructure.
• Scalability: The techniques can be applied to large-scale networks, making it suitable for various industries and deployments.

Abstract

Techniques are disclosed for intelligent infrastructure management in a radio access network. For example, a method obtains, from a plurality of baseband units of a radio access network, a plurality of data sets, wherein respective ones of the plurality of data sets correspond to operational statistics of respective ones of the plurality of baseband units. The method then generates a forecasted data set corresponding to one or more predicted operational statistics of each of the subset of baseband units, wherein the forecasted data set is generated using a first machine learning model. The method analyzes the forecasted data set to predict a future occurrence of a technical issue and to determine at least one corrective action for the predicted future occurrence of the technical issue, wherein the analysis is performed using a second machine learning model. The method then causes initiation of the at least one corrective action.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR INFORMATION-CENTRIC NETWORKING (17858670)

Main Inventor

Zijia Wang

Brief explanation

The abstract describes a method, electronic device, and computer program for information-centric networking. The method involves using a memory layer in a machine learning model to obtain future information based on an environmental state obtained from information-centric networking at a future moment. The machine learning model is then trained using this future information to improve the cache mechanism in information-centric networking.

• The method uses a memory layer in a machine learning model to obtain future information in information-centric networking.
• The machine learning model is trained using the future information to improve the cache mechanism.
• This approach allows for more efficient information-centric networking based on reinforcement learning.

Potential Applications

• This technology can be applied in various information-centric networking systems to improve cache mechanisms.
• It can enhance the performance and efficiency of content delivery networks (CDNs) by predicting future information needs.
• It can be used in smart cities to optimize data caching and improve network performance.

Problems Solved

• Traditional information-centric networking systems may not efficiently utilize caching mechanisms.
• Predicting future information needs in information-centric networking can be challenging.
• This technology solves these problems by using a machine learning model and future information to improve the cache mechanism.

Benefits

• The use of future information in training the machine learning model improves the efficiency of information-centric networking.
• It allows for better prediction of future information needs, leading to more effective caching.
• This technology can optimize network performance, reduce latency, and enhance user experience.

Abstract

Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for information-centric networking. In the method, a memory layer in a machine learning model is used to obtain, on the basis of an environmental state obtained from information-centric networking at a future moment, future information associated with a memory layer corresponding to the future moment, and the machine learning model is trained using the future information. By means of the solution, a model trained using future information can be obtained. By use of the model, information-centric networking based on reinforcement learning achieves a more efficient cache mechanism.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR CROSS-REGIONAL DATA SEARCHING (17855884)

Main Inventor

Zhenzhen Lin

Brief explanation

The present disclosure describes a method, electronic device, and computer program product for cross-regional data searching. The method involves the following steps:

1. Acquiring a data identifier of the target data when a searching request for the target data is received at a first node. 2. Determining a second node in the same region as the first node that stores metadata of the target data, based on the data identifier. The metadata includes the data identifier. 3. Determining a third node in a different region from the first region that stores the actual target data, based on the metadata.

• The method allows for efficient searching of data across different regions.
• It enables the identification of the location of the target data based on its data identifier.
• The method ensures that the actual target data is stored in a node located in a different region from the requesting node.

Potential applications of this technology:

• Cloud computing: The method can be used to search for and access data stored in different regions of a cloud infrastructure.
• Distributed databases: It can facilitate the retrieval of data stored in different regions of a distributed database system.
• Content delivery networks: The method can help in locating and retrieving content stored in different regions of a content delivery network.

Problems solved by this technology:

• Efficient cross-regional data searching: The method provides a systematic approach to search for data stored in different regions, eliminating the need for manual searching or complex algorithms.
• Data accessibility: It ensures that data can be accessed from any region, regardless of the location of the requesting node.
• Scalability: The method can handle a large volume of data stored in different regions, allowing for scalability in data storage and retrieval.

Benefits of this technology:

• Improved efficiency: The method streamlines the process of searching for data across different regions, saving time and resources.
• Enhanced data accessibility: It enables users to access data from any region, promoting seamless data sharing and collaboration.
• Scalability and flexibility: The method can be applied to various systems and infrastructures, accommodating the storage and retrieval of data in different regions.

Abstract

Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for cross-regional data searching. The method includes acquiring a data identifier of target data in response to receiving a searching request for the target data at a first node. The method includes determining, based on the data identifier, a second node storing metadata of the target data, wherein both the first node and the second node are located in a first region, and the metadata includes the data identifier. The method further includes determining, based on the metadata, a third node storing the target data, wherein the third node is located in a second region different from the first region.

INFRASTRUCTURAL EDGE SECURITY AS A SERVICE (17806635)

Main Inventor

Victor Fong

Brief explanation

==Abstract==

The patent application discloses edge security for distributed infrastructure. Hosts in the infrastructure have data processing units that can be centrally managed. Security policies can be selectively distributed and localized at the hosts, allowing for real-time reactions to security violations. The security policies can be rule-based or inference mode-based.

Bullet Points

• Distributed infrastructure equipped with data processing units
• Central management of hosts using a control plane
• Selective distribution and localization of security policies
• Real-time reactions to security violations
• Rule-based and inference mode-based security policies

Potential Applications

• Cloud computing environments
• Internet of Things (IoT) networks
• Industrial control systems
• Edge computing platforms

Problems Solved

• Lack of centralized management for distributed infrastructure
• Inability to react to security violations in real-time
• Difficulty in distributing and localizing security policies

Benefits

• Enhanced security for distributed infrastructure
• Real-time reactions to security violations
• Centralized management for improved efficiency
• Flexibility in choosing security policy types

Abstract

Edge security is disclosed. Hosts included in a distributed infrastructure are equipped with data processing units that may be centrally managed using a control plane. Security policies can be selectively distributed to the hosts and localized at the hosts. Reactions to security violations can be initiated at the hosts in substantially real-time. The security policies may take various forms, including rule-based security policies and inference mode-based security policies.

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR VIDEO PROCESSING (17857739)

Main Inventor

Qiang Chen

Brief explanation

The present disclosure describes a method, electronic device, and computer program product for video processing. The method involves receiving a low-resolution video from a first device, converting it into a high-resolution video using a computing model associated with the user, and sending the high-resolution video to a second device. This approach reduces bandwidth usage and improves video quality.

• The method involves receiving a low-resolution video from a first device.
• The low-resolution video includes an image of a user.
• The method converts the low-resolution video into a high-resolution video.
• The conversion is based on a computing model associated with the user.
• The high-resolution video has a higher resolution than the original video.
• The method sends the high-resolution video to a second device.
• This approach reduces the need for acquiring and uploading high-resolution videos.
• Bandwidth usage is minimized as only low-resolution videos are transmitted.
• The quality of the reconstructed videos is improved using the computing model.

Potential Applications

• Video conferencing and communication platforms.
• Live streaming services.
• Video sharing platforms.
• Online gaming platforms.
• Virtual reality applications.

Problems Solved

• Reduces bandwidth usage by transmitting low-resolution videos.
• Improves video quality by reconstructing high-resolution videos using a computing model.
• Minimizes the need for acquiring and uploading high-resolution videos.

Benefits

• Saves bandwidth and reduces data usage.
• Enhances the quality of videos by providing high-resolution reconstructions.
• Provides a more efficient and effective video processing method.
• Improves user experience in video communication and streaming.

Abstract

Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for video processing. The method for video processing includes receiving a first video with a first resolution from a first terminal device, the first video including an image of a first user. The method further includes converting the first video into a second video with a second resolution based on a computing model associated with the first user, the second resolution being higher than the first resolution. The method further includes sending the second video to a second terminal device. In this manner, only low-resolution videos need to be acquired and uploaded, and high-resolution videos can be obtained through reconstruction by using a computing model customized for video services, thereby ensuring that less bandwidth is occupied and improving the quality of the reconstructed videos.

USER PRESENCE BASED MEDIA MANAGEMENT (17861922)

Main Inventor

Kai Wang

Brief explanation

An information handling system can detect media playback and determine the user's presence based on data from a user presence device. It can then adjust a setting associated with the media playback based on the user's presence.

• The system detects media playback.
• It collects data from a user presence device.
• It determines the user's presence relative to the information handling system using the data from the user presence device.
• The system applies a policy based on the user's presence.
• It adjusts a setting associated with the media playback based on the user's presence.

Potential Applications

• Smart home systems that adjust media playback settings based on user presence.
• Video conferencing systems that optimize settings based on the presence of participants.
• Public display systems that adjust media playback based on the number of people present.

Problems Solved

• Inefficient use of media playback settings when users are not present.
• Lack of automation in adjusting media playback settings based on user presence.
• Difficulty in optimizing media playback settings for different scenarios.

Benefits

• Improved energy efficiency by adjusting media playback settings when users are not present.
• Enhanced user experience by automatically optimizing media playback settings based on user presence.
• Increased convenience by eliminating the need for manual adjustment of media playback settings.

Abstract

An information handling system may detect media playback, collect data from a user presence device, and determine user presence relative to the information handling system based on the data from the user presence device. The system may also apply a policy based on the user presence by adjusting a setting associated with the media playback.

AUTOFOCUS ACCURACY AND SPEED USING THERMAL INPUT INFORMATION (17837257)

Main Inventor

Yi Hsien Lin

Brief explanation

The patent application describes an imaging device that includes an image sensor, lens assembly, temperature sensor, and processor. The lens assembly has an adjustable distance from the image sensor. The temperature sensor determines the temperature of the device, and the processor uses this information to select an auto focus table and provide auto focus for the object.

• The imaging device has an adjustable lens assembly that can change the distance between the lens and image sensor.
• A temperature sensor is included to determine the temperature of the device.
• The processor uses the temperature information to select an auto focus table.
• The auto focus table provides a correlation between the object distance and the stroke distance of the lens assembly.
• The auto focus table is temperature compensated, meaning it takes into account the temperature of the device.
• The processor uses the selected auto focus table to provide accurate auto focus for the object.

Potential Applications

• Photography and videography equipment
• Surveillance cameras
• Medical imaging devices
• Industrial inspection systems

Problems Solved

• Inaccurate auto focus due to temperature variations
• Ensuring consistent and reliable auto focus performance in different temperature conditions

Benefits

• Improved accuracy and reliability of auto focus
• Temperature compensation ensures consistent performance
• Enhanced image quality and clarity

Abstract

An imaging device includes an image sensor, a lens assembly, a temperature sensor, and a processor. The lens assembly includes a lens whose distance from the image sensor is adjustable from a minimum stroke distance to a maximum stroke distance. The temperature sensor determines a temperature of the imaging device. The processor instantiates auto focus tables that are each correlated to a temperature, receives at temperature from the temperature sensor, selects an auto focus table associated with the temperature, and provides an auto focus for the object based upon the auto focus table. Each auto focus table provides a temperature compensated correlation of a first stroke distance associated with the object at the maximum distance from the imaging device, and a second stroke distance associated with the object at the minimum distance from the imaging device.

MULTIFOCAL CAMERA BY REFRACTIVE INSERTION AND REMOVAL MECHANISM (17837551)

Main Inventor

Yi-Hsien Lin

Brief explanation

The patent application describes an information handling system camera with a rapid focus capability for near and far objects. This is achieved by selectively inserting and removing a refractive material between the camera lens and image sensor to adjust the focusing distance.

• The camera uses a refractive material to adjust the focusing distance between the lens and image sensor.
• When focusing on far objects, a refractive material with specific properties is inserted to achieve instant focus.
• For near objects, the refractive material is removed or replaced with a thinner material to quickly achieve focus.
• A voice coil motor or other actuator can be used to fine-tune the focus when needed.
• In an alternative embodiment, focus is solely achieved by inserting and retracting the refractive material, eliminating the need for a lens actuator.

Potential Applications

• This technology can be applied to various information handling system cameras, such as smartphones, tablets, or digital cameras.
• It can improve the speed and accuracy of focusing in photography and videography applications.
• The rapid focus capability can be beneficial in capturing fast-moving objects or in situations where quick adjustments in focus are required.

Problems Solved

• The technology solves the problem of slow focusing times associated with traditional lens actuator-based focus systems.
• It provides a rapid focus capability for both near and far objects, enhancing the overall user experience in capturing images or videos.

Benefits

• The use of a refractive material allows for instant focus adjustment, reducing the time required to achieve focus.
• The ability to quickly switch between near and far object focus improves the versatility and efficiency of the camera.
• The elimination of a lens actuator in the alternative embodiment simplifies the camera design and potentially reduces manufacturing costs.

Abstract

An information handling system camera that includes a rapid focus capability for near and far objects by selectively inserting and removing a refractive material between the camera lens and image sensor to effectively adjust the camera focusing distance. When far object focus is desired, a refractive material of predefined properties is inserted to enable an instant focus of the far object to the image sensor. When a near object focus is desired, the refractive material is removed (or replaced with a thinner material) to rapidly achieve focus. In addition, a voice coil motor or other actuator fine tunes the focus when desired. In one alternative embodiment, focus is performed only with the insertion and retraction of the refractive material so that no lens actuator is included in the camera. The refractive material extends the back focal distance between the lens and image sensor relative to the back focal distance without the refractive material. Rapid insertion of refractive material shortens the time to adjust focus relative to focus time associated with actuator-based lens movement focus.