4423 results about "Computational resource" patented technology

In language evaluation systems, user expressions are often evaluated by speech recognizers and language parsers, and among several possible translations, a highest-probability translation is selected and added to a dialogue sequence. However, such systems may exhibit inadequacies by discarding alternative translations that may initially exhibit a lower probability, but that may have a higher probability when evaluated in the full context of the dialogue, including subsequent expressions. Presented herein are techniques for communicating with a user by formulating a dialogue hypothesis set identifying hypothesis probabilities for a set of dialogue hypotheses, using generative and/or discriminative models, and repeatedly re-ranks the dialogue hypotheses based on subsequent expressions. Additionally, knowledge sources may inform a model-based with a pre-knowledge fetch that facilitates pruning of the hypothesis search space at an early stage, thereby enhancing the accuracy of language parsing while also reducing the latency of the expression evaluation and economizing computing resources.

Embodiments of the invention address deficiencies of the art in respect to dynamic computing resource allocation, and provide a method, system and computer program product for dynamic resource allocation for disparate application performance requirements. In one embodiment of the invention, a resource allocation data processing system can include a shared resource pool including resources and a resource configurator coupled to the shared resource pool. The system further can include a service processor coupled to the resource configurator, wherein the service processor can include an application programming interface (API) exposing methods for commanding the resource configurator to configure the resources in the shared resource pool.

An adaptive compression mechanism that dynamically selects compression algorithms applied to network application traffic to improve performance. One implementation includes an arbitration scheme that reduces the impact on computing resources required to analyze different compression algorithms for different network applications. The adaptive compression functionality of the present invention can be integrated into network application traffic management or acceleration systems.

Data is collected relating to the availability of a computing resource such as data identifying a start event that indicates that the computing resource became available for use and data identifying a stop event that indicates that the computing resource became unavailable for use. The data relating to the availability of the computing resource is processed to determine the outage time for the computing resource during a period of time, such as a 24-hour period. The outage time is then utilized to calculate the level of availability for the computing resource during the period of time.

A hybrid of grid-based and graph-based search computations, together with provision of a sparse search technique effectively limited to high-probability candidate nodes provides accommodation of path constraints in an optimization search problem in substantially real-time with limited computational resources and memory. A grid of best cost (BC) values are computed from a grid of map cost (MC) values and used to evaluate nodes included in the search. Minimum segment/vector length, maximum turn angle, and maximum path length along a search path are used to limit the number of search vectors generated in the sparse search. A min-heap is preferably used as a comparison engine to compare cost values of a plurality of nodes to accumulate candidate nodes for expansion and determine which node at the terminus of a partial search path provides the greatest likelihood of being included in a near-optimal complete solution, allowing the search to effectively jump between branches to carry out further expansion of a node without retracing portions of the search path. Capacity of the comparison engine can be limited in the interest of expediting of processing and values may be excluded or discarded therefrom. Other constraints such as approach trajectory are accommodated by altering MC and BC values in a pattern or in accordance with a function of a parameter such as altitude or by testing of the search path previously traversed.

A system capable of dynamically reserving bandwidth and adjusting bandwidth reservations for active sessions of data communication in a data communications device is provided. The system generally separates the operation of bandwidth allocation and adjustment from the operation of data transport through the device, thereby allowing bandwidth reservations and adjustments to be made without disturbing sessions of data communication that are actively being transported through the device. The system can accept requests to allocate or reserve bandwidth in a data communications device using bandwidth reservation protocols such as RSVP. The reservation requests create sender state data that can be used to compute resource allocation data. The resource allocation data can be used to label data storage locations in a data storage mechanism according to the required bandwidth reservations. A data scheduling apparatus, which is ignorant of particular sessions and specific amounts of reserved bandwidth, examines data and deposits data into data storage locations having a label corresponding to a session identification specified in the data, if any. If an unknown or no session identification is specified in the data, the data scheduler deposits data into a data storage location that is unlabeled or that has an unreserved label. Thus session bandwidth is determined by the percentage of labeled data storage locations for the session. Changes in bandwidth reservations are reflected in the separate operation of alterations made in the data storage labeling scheme, and do not affect the data scheduler, or data dequeuing mechanisms, thus allowing data sessions to continue without interruption during bandwidth adjustments.

Improved approaches to flexibly implementing graphic equalizers on media players are disclosed. These approaches provide dynamic order reduction of a multi-band graphic equalizer so that equalizer effects can be timely performed with only limited computational resources. In one embodiment, a media player receives a media item and associated equalizer settings for a multi-band graphic equalizer. The media player can then automatically (i.e., without user action) approximate the multi-band graphic equalizer with the equalizer settings for the media item using a fewer number of filters. Fewer filters means order reduction, and thus reduction in computational requirements. After the multi-band graphic equalizer is approximated, the media player can present the media item to its user in accordance with the reduced complexity, approximated equalizer.

A cloud enablement aggregation proxy (CEAP) receives and processes audit data from audited resources before such data is stored in a database. The CEAP manages log data for resources hosted in a multi-tenant shared pool of configurable computing resources (e.g., a compute cloud). A method for managing log data begins by the proxy aggregating and normalizing log information received from a plurality of the resources. The aggregated and normalized log information is then parsed to identify a tenant associated with each of a set of transactions. For each of the set of transactions, the CEAP annotates log data associated with the tenant and the particular transaction to include a tenant-specific identifier. An optional tenant separation proxy (TSP) separates the annotated log data on a per tenant basis prior to storage, and the tenant-specific log data may be stored in per tenant data structures or dedicated tenant log event databases to facilitate subsequent compliance or other analysis.

A realtime kernel supports realtime communications between communicants operating on respective network nodes. The realtime kernel handles the complex tasks of connecting to communicants, virtual areas, and other network resources, switching those connections in response to user inputs, and mixing realtime data streams. The realtime kernel enables developers to focus on developing high-level communications functionality instead of low-level plumbing code. The realtime kernel imposes relatively low computational resource requirements so that realtime communications performance can be achieved using a wide range of computing devices and network connections that currently are available.

Systems and methods implementing a demand-driven workload scheduling optimization of shared resources used to execute tasks submitted to a computer system are disclosed. Some embodiments include a method for demand-driven computer system resource optimization that includes receiving a request to execute a task (said request including the task's required execution time and resource requirements), selecting a prospective execution schedule meeting the required execution time and a computer system resource meeting the resource requirement, determining (in response to the request) a task execution price for using the computer system resource according to the prospective execution schedule, and scheduling the task to execute using the computer system resource according to the prospective execution schedule if the price is accepted. The price varies as a function of availability of the computer system resource at times corresponding to the prospective execution schedule, said availability being measured at the time the price is determined.

Web services interface policy constraints may be specified in a policy constraints language and policy processing, such as generating an intersection policy of two policies may be automated by a policy-processing engine. A policy constraint may be a specification of a value, range of values, or set of values that a particular requirement or offering is allowed to have. Hierarchies of requirements and/or offerings may also be expressed and matched such that a more specific case of a requirement or offering may be matched against a more general case of the same requirement or offering. Also, preferences among vocabulary items, vocabulary item values, policy constraints, and other elements of a policy may be specified and automatically determined by a policy-processing engine. Automated matching of consumer requirements against provider offerings may allow a policy-processing engine to process policies with specifications of requirements or offerings from any domain-specific schema.

The invention relates to a cloud computing resource scheduling method based on dynamic reconfiguration virtual resources. The method comprises the steps of: using cloud application load information collected by a cloud application monitor as a basis; making a dynamic decision based on the load capacity of the virtual resources for running cloud application and the current load of the cloud application; and dynamically reconfiguring virtual resources for cloud application based on the decision result. Dynamic adjustment of resources is realized by a method for reconfiguring virtual resources for cloud application, without needing dynamic redistribution of physical resources or stopping executing cloud application. The method can dynamically reconfigure the virtual resources according to the load variation of the cloud application, optimize allocation of the cloud computing resources, realize effective use of the cloud computing resources, and meet the requirements on dynamic scalability of cloud application. In addition, the method can avoid waste of the cloud computing resources, and save the cost for using resources for cloud application users.

A dynamically configurable resource pool may provide a pool of computing resource for use in a computing system or application, such as a connection pool or a thread pool for server systems such as application and web server systems. In one embodiment, a server may include a resource pool configured to provide a plurality of computing resources. Other components in the server may be configured to request use of one of the computing resources from the connection pool. The resource pool may include a resource pool manager configured to service requests for the computing resources. The resource pool manager may manage configuration of the resource pool. The resource pool manager may also be configured to receive a configuration change request to change the configuration of the resource pool while the resource pool is available for use.

The invention provides an off-the-shelf product solution to target the specific needs of commercial users with naturally parallel applications. A top-level, public API provides a simple “compute server” or “task farm” model that dramatically accelerates integration and deployment. A number of described and claimed adaptive scheduling and caching techniques provide for efficient resource and/or network utilization of intermittently-available and interruptible computing resource in distributed computing systems.

The invention discloses a joint optimization method for task unloading and resource allocation in a mobile edge computing network, which comprises the following steps of 1, establishing an OFDMA (Orthogonal Frequency Division Multiple Access)-based multi-MEC (Mobile Edge Computing) base station and a multi-user scene model, wherein the MEC base station supports the multi-user access; 2, introducing an unloading decision mechanism; Meanwhile, constructing a local calculation model and a remote calculation model, selecting a user needing to perform calculation unloading, and establishing a calculation task unloading and resource allocation scheme based on minimum energy consumption under the condition of meeting the time delay constraint according to the conditions; 3, carrying out variablefusion on three mutually constrained optimization variables, namely an unloading decision variable, a wireless resource distribution variable and a computing resource distribution variable, so as to simplify the problem; and 4, obtaining an unloading decision and a resource allocation result which enable the total energy consumption of the user in the MEC system to be minimum through a branch andbound algorithm. The method has the advantage that the energy consumption of the system can be effectively reduced on the premise that strict time delay limitation is guaranteed.

A closed-loop vision system is disclosed that utilizes a concept known as Dynamically Reconfigurable Vision (DRV), which is adaptive image sensing driven by a computer or human operator's response to changing scenery. The system reduces the amount of irrelevant video information sensed and thus achieves more effective bandwidth and computational resource utilization, as compared to traditional vision systems. One or more reconfigurable photodetector arrays sensitive to either visible, infrared or ultraviolet radiation are present in the DRV system. These photodetector arrays feature on-chip means for spatial and temporal data reduction implemented through multiple independently controllable, time-correlated, frequently overlapping windows on the photodetector array that may be programmed according to their size, location, resolution, integration time, and frame rate. All photodetector array windows are dynamically reconfigurable in real time on a frame-by-frame basis. Furthermore, a DRV system is constructed in a client-server architecture in which a vision processor client passes window request command messages to the reconfigurable photodetector array server, which in turn delivers the requested video back to the client processor. The ability to simultaneously reconfigure, integrate, process, and readout multiple photodetector array video windows is an important characteristic of the DRV system.

Generally speaking, systems, methods and media for management of grid computing resources based on service level requirements are disclosed. Embodiments of a method for scheduling a task on a grid computing system may include updating a job model by determining currently requested tasks and projecting future task submissions and updating a resource model by determining currently available resources and projecting future resource availability. The method may also include updating a financial model based on the job model, resource model, and one or more service level requirements of an SLA associated with the task, where the financial model includes an indication of costs of a task based on the service level requirements. The method may also include scheduling performance of the task based on the updated financial model and determining whether the scheduled performance satisfies the service level requirements of the task and, if not, performing a remedial action.

Methods and systems for improved management of power utilization and resource consumption among physical hosts in a cloud computing environment. The management server may provide functionality facilitating the identification and optimized placement of a virtual machine within a cloud computing environment by evaluating historical and heuristic metrics data associated with both the physical hosts and the virtual machines. The management server utilizes the metrics data to generate scores for a plurality of physical host based on physical resources available in a cloud of computing resources. The management server identifies a physical host on which to place a virtual machine using the metrics data, generated scores, and numerous, configurable criteria. The management server responds to the identification of the physical host on which to place a virtual machine by adjusting processor performance and/or operating states for one or more of the physical hosts in the cloud computing environment.

A method for operating a cluster of N server nodes to service client requests received on a network. Each client request is directed to one of C customers hosted on the server cluster. Each customer is identified by a domain name, and each server node is identified by an address on a network. In the method of the present invention, the customers are grouped into N groups, each group being assigned to a corresponding one of the server nodes. Configuration information is provided to a Domain Name Server (DNS), the information defining the correspondence between each of the customers and one of the server nodes assigned to one of the groups containing that customer. The DNS provides the address of the server node in response to a message specifying the domain name of the customer. The client then directs its request to the identified server node utilizing the address provided by the DNS. In the preferred embodiment of the present invention, the grouping of the customers depends on a measurement of the computational resources required to service the client requests for each of the customers. In embodiments in which the activity associated with each request is primarily the return of files stored in the cluster, the measurement of computational resources includes the size of the files returned by each client—within a time period and the communication bandwidth needed to service the requests.

Techniques for creating a High Dynamic Range (HDR) image within a consumer grade digital camera from a series of images of a scene captured at different exposure levels, and displaying the HDR image on the camera's built-in display are provided. The approach employs mixing images of the series to incorporate both scene shadow and highlight details, and the removing of “ghost” image artifacts appearing in the mixed HDR image resulting from movement in the scene over the time the series images are captured. The low computational resource utilization of the image mixing and ghost removal processing operations, along with the ability to commence image mixing and ghost removal prior to the acquisition of all series images, can significantly reduce the time required to generate and display a tone mapped HDR image.