107 results about "Row-major order" patented technology

Techniques are disclosed for converting data into a format tailored for efficient multidimensional fast Fourier transforms (FFTS) on single instruction, multiple data (SIMD) multi-core processor architectures. The technique includes converting data from a multidimensional array stored in a conventional row-major order into SIMD format. Converted data in SIMD format consists of a sequence of blocks, where each block interleaves s rows such that SIMD vector processors may operate on s rows simultaneously. As a result, the converted data in SIMD format enables smaller-sized 1D FFTs to be optimized in SIMD multi-core processor architectures.

Methods for linking multiple events involving firearms submit event information data and firearms information data to a system that includes a database for multiple firearms events and generate a query map that correlates data in the system database with a preselected geographical grid based upon one or more preselected matching criteria to one or more data entries for a selected event. The map is used to generate a request for a possible link analysis which may be initiated by a firearms examiner and is then performed to determine if there is a link. The event information data can include a case number, a geographical location data entry, a date entry, a crime type, a weapon involved and a suspect information data field based upon input from an investigating officer while the firearms information data can include a weapon data entry and a bullet data entry based upon input from a person with firearms examiner training. The methodology can be used to prioritize multiple requests for a possible link analysis based upon information obtained in one or more query responses and the query response can display at least some of its information in a query map that correlates data in the system database with a preselected geographical grid. One or more gang database modules allow users to execute searches of a gang database and plot incidents on a geographic grid with each incident being represented by a unique icon related to a class of crime or type of incident.

A method of reducing spatial noise in an image. Low-pass (smoothing) filters are calculated simultaneously from three successive image rows. Three blocks (m1, m2, m3) are associated with the three successive image rows, and the blocks are processed in row-major order. This implementation is applicable to both luminance and chrominance. The number of smoothing parameters is reduced to one. The technique is applicable to both luminance and chrominance. Directional mapping is used. Extension of the technique to spatial filtering using a 5×5 neighborhood (using five successive image rows) is described. Embodiments of the method using the MMX instruction set are described.

A system and method for prioritizing contacts is disclosed. The method discloses: receiving a set of requests from a set of contacts to be connected to a recipient; generating a first priority level score for each request based on an acoustical analysis of a contact speech signal within the request; generating a second priority level score for each request based on a keyword analysis of the request; combining the priority level scores for each request into a combined priority level score; prioritizing the requests within the set of requests based on their respective combined priority level scores; and routing to the recipient that request, from the set of requests, having a highest combined priority level score. The system of the present invention includes all means and mediums for implementing the method.

Improved caching of content at caching proxy (“CP”) servers is disclosed. In one aspect, negotiations occur before content is dynamically distributed, whereby an entity such as a Web server selects content and at least one target CP server, and sends a content distribution request to each target, describing the content to be distributed. Preferably, the selection is made by dynamically prioritizing content based on historical metrics. In another aspect, a CP server that receives a content distribution request during these negotiations determines its response to the distribution request. Preferably, content priority of already-cached content is compared to priority of the content described by the content distribution request when making the determination. In yet another aspect, a CP server selectively determines whether to cache content during cache miss processing. Preferably, this comprises comparing content priority of already-cached content to priority of content delivered to the CP server during the cache miss.

Methods, systems and data structures select prioritized robust data values from a plurality of available data values formed by a plurality of data bits, each capable of exhibiting a bit value. Available data values are arranged into a gray code format, and alternate values of gray code format are selected to form a value map. An optional complementary value map may also be formed from the remaining data values. The value map is then prioritized according to bit adjacencies, wherein bit adjacencies are defined by contiguous bits within one of the data values that exhibit a common bit value. Priority may be given to data values having shortest and / or fewest bit adjacencies.

An apparatus and method utilizes system memory as backing stores so that local graphics memory may be oversubscribed. Surfaces may be paged in and out of system memory based on the amount of usage of the surfaces. The apparatus and method also prioritizes surfaces among different tiers of local memory (e.g. frame buffer), non-local memory (e.g. page locked system memory), and system memory backing stores (e.g. pageable system memory) locations based on predefined criteria and runtime statistics relating to the surfaces. As such, local memory may be, for example, expanded without extra memory costs such as adding a frame buffer memory to allow graphics applications to effectively use more memory and run faster.

An apparatus for scheduling dispatch of instructions among a plurality of threads being concurrently executed in a multithreading processor is provided. The apparatus includes an instruction decoder that generate register usage information for an instruction from each of the threads, a priority generator that generates a priority for each instruction based on the register usage information and state information of instructions currently executing in an execution pipeline, and selection logic that dispatches at least one instruction from at least one thread based on the priority of the instructions. The priority indicates the likelihood the instruction will execute in the execution pipeline without stalling. For example, an instruction may have a high priority if it has little or no register dependencies or its data is known to be available; or may have a low priority if it has strong register dependencies or is an uncacheable or synchronized storage space load instruction.

A computer system and method is disclosed for mining current and archived address data in order to identify a preferred address for each service point in a territory. The data mining system may start in response to the presentation of a candidate address for matching. The set of mined data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms with nodes joined according to common characteristics. A computer system and method for maintaining a central database of preferred addresses is also disclosed. Selected address data gathered in a queue may be scored by characteristic, grouped by consignee location, and staged for processing. The scored queue of data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms.