86results about How to "High degree of parallelism" patented technology

In a memory system having multiple erase blocks in multiple planes, a selected number of erase blocks are programmed together as an adaptive metablock. The number of erase blocks in an adaptive metablock is chosen according to the data to be programmed. Logical address space is divided into logical groups, a logical group having the same size as one erase block. Adaptive logical blocks are formed from logical groups. One adaptive logical block is stored in one adaptive metablock.

Data may be stored in a non-volatile memory array in adaptive metablocks that are configured according to the locations of data boundaries in the data. Data may be stored in an intermediate format and later copied to adaptive metablocks configured for the data. Data in intermediate format may be stored in non-volatile random access memory or in a portion of the non-volatile memory array.

Data in data runs are stored in a non-volatile memory array in adaptive metablocks that are configured according to the locations of data boundaries. A serial flash buffer is used to store some data, while other data are directly stored in non-volatile memory. Data may be stored with alignment to data boundaries during updating of the data to improve efficiency of subsequent updates.

An instruction scheduler in an optimizing compiler schedules instructions in a computer program by determining the lifetimes of fixed registers in the computer program. By determining the lifetimes of fixed registers, the instruction scheduler can achieve a schedule that has a higher degree of parallelism by relaxing dependences between instructions in independent lifetimes of a fixed register so that instructions can be scheduled earlier than would otherwise be possible if those dependences were precisely honored.

An integrated circuit 4 is provided including an array 10 of processors 26 with interface circuitry 12 providing communication with further processing circuitry 14. The processors 26 within the array 10 execute individual programs which together provide the functionality of a cycle-based program. During each program-cycle of the cycle based program, each of the processors executes its respective program starting from a predetermined execution start point to evaluate a next state of at least some of the state variables of the cycle-based program. A boundary between program-cycles provides a synchronisation time (point) for processing operations performed by the array.

The invention discloses a real-time transparent object GPU (graphic processing unit) parallel generating method based on three-dimensional point cloud, which comprises steps of: (1) generating a background map of a nontransparent object; (2) generating geometric buffer of a transparent object, generating all three-dimensional points by a sphere generating way, utilizing hardware depth detection to acquire depth of an approximate plane, and simultaneously saving material information of the transparent object; (3) smoothing the depth, using the depth information in the geometric buffer to perform smooth filtering to the depth to acquire a smooth surface; (4) calculating thickness of the transparent object, generating all three-dimensional points, and utilizing hardware Alpha to mix and calculate the thickness of the transparent object; and (5) coloring the transparent object, using the depth and the material information to perform illumination computation to the transparent object, utilizing the thickness to calculate refraction and reflection properties of the transparent object, and utilizing the background map to finish coloring. The method avoids the surface rebuilding step in the traditional method, and can meet the requirement of real-time generation of a million-level transparent object based on point cloud.

For switching or transmitting data packets, one can provide communication systems which consist of several modules—operating in parallel on segments of a packet—to increase speed and handling capacity. One module acts as master (21), the others are slave modules (22) controlled by control signals (25) derived by the master module. It is important that in each module the data segment and the respective control signal of each packet are correctly synchronized, because in large systems the data paths carrying packet segments and the control signal paths may have substantially different delays. The invention provides for measurement of the propagation delay differences and for introducing a controlled delay in each slave module, so that data segments and control signals can be correctly correlated by delaying either the one or the other. Synchronization packets are transmitted besides normal data packets, for obtaining time stamps which are used to determine the delay difference.

A method for producing a multi-chip module having application of at least one contact elevation onto a substrate, application and patterning of a rewiring device onto the substrate and the at least one contact elevation with provision of a contact device on the at least one contact elevation, application of a semiconductor chip onto the substrate with electrical contact-connection of the rewiring device; application of an encapsulating device that is not electrically conductive onto the semiconductor chip, the substrate, the rewiring device and the at least one contact elevation, the contact device on the at least one contact elevation at least touching a first surface of the encapsulating device; and repetition at least once of at least the first two steps, the first surface of the encapsulating device serving as a substrate and the correspondingly produced rewiring device making electrical contact with the contact device of the at least one contact elevation of the underlying plane.

A microfluidic device (1000-1005), comprising: a semiconductor body (2) having a first side (2a) and a second side (2b) opposite to one another, and housing, at the first side, a plurality of wells (4), having a first depth; an inlet region (30) forming an entrance point for a fluid to be supplied to the wells; a main channel (6a) fluidically connected to the inlet region, and having a second depth; and a plurality of secondary channels (6b) fluidically connecting the main channel to a respective well, and having a third depth. The first depth is higher than the second depth, which in turn is higher than the third depth. According to an aspect, the microfluidic device further comprises a cover layer (8), arranged above the first side of the semiconductor body, configured for sealing the wells and provided with at least a first valve hole (54) which extends through the cover layer and overlaps, at least partially, the secondary channels; and a flexible layer (14), arranged above the cover layer and provided with at least a protrusion (74) extending through the first valve hole towards the semiconductor body and overlapping, at least partially, the secondary channels, the flexible layer being configured such that, when a pressure is applied on it, the protrusion contacts the semiconductor body and enters the secondary channels thus fluidically isolating the wells from one another.

The invention provides a method and a system for carrying out matrix product operation on a computer cluster. A distributed file system and a plurality of calculation nodes for executing Map tasks and Reduce tasks are arranged on the computer cluster. The method specifically comprises the following step of: executing a first Map task, a first Reduce task, a second Map task and a second Reduce task on the calculation nodes, wherein the first Map task is used for carrying out first treatment to obtain a corresponding first key value pair result; the first Reduce task is used for gathering the key value of the same main key in the first key value pair result; the second Map task is used for identifying to obtain elements of a first matrix and a second matrix, and carrying out two-two combining and multiplying operation to obtain a second key value pair result; and the second Reduce task is used for summing the key values of the same main key in the second key value pair result. According to the method and the system provided by the invention, the calculation speed of the matrix product can be improved.

The present invention belongs to the field of the geographic space information processing technology, and relates to a parallel mode grid image slicing method. The method provided by the invention comprises the following steps: the step 1: obtaining an original grid image, and setting a target tile grade and the total number of processes; the step 2: reading meta-data information of the original grid image; the step 3: converting the original grid image to WebMercator projection; the step 4: calculating the max grade and the mix grade of the tiles of the projection conversion result image; the step 5: calculating the range of the tile rank numbers of the projection conversion result image, and establishing an empty tile file; and the step 6: dividing tasks for each process through adoption of a wheel method; the step 7: reading tiles in each task pool one by one through each process, inversely calculating the geographical range with respect to the tiles, calculating the relative position and the size of an intersecting region in the projection conversion result image, and reading the original grid data of the intersecting region into a memory space; and the step 8: writing resampling data in built empty tile file after resampling the original grid data of the intersecting region.

This invention relates to one agreement and system for network memory in computer information memory technique field by use of VI composed of start process and aim process. The invention system comprises the following parts: one start device and N aim parts connected to the internet through VI network interface, VI connection channel; the start and aim parts are connected to the channel transmission data through more than one VI.

A non-volatile memory device includes a chip of semiconductor material. The chip includes a memory and control means for performing a programming operation, an erasing operation and a reading operation on the memory in response to corresponding external commands. The chip further includes testing means for performing at least one test process including the repetition of at least one of said operations by the control means, and a single access element for enabling the testing means.

A patient interface device includes: a first interface port configured to be interfaced with a laser surgery apparatus; a second interface port configured to be interfaced with a patient's eye, the second interface port including an applanating lens for application to a patient's eye during a laser surgery procedure; a chamber extending between the first interface port and the second interface port and defining a chamber therein, wherein air may be evacuated from the chamber by the laser surgery apparatus via the first interface port; and a tubular light guiding structure having at a first end thereof a light receiving surface, configured to receive light, and having at a second end thereof a light-emitting surface, wherein the second surface is disposed adjacent the applanating lens and configured to provide the light in a vicinity of the patient's eye when the applanating lens is applied to the patient's eye.

A digital computing component and method for computing configured to execute the constructs of a high-level software programming language via optimizing hardware targeted at the particular high-level software programming language. The architecture employed allows for parallel execution of processing components utilizing instructions that execute in an unknown number of cycles and allowing for power control by manipulating the power supply to unused elements. The architecture employed by one or more embodiments of the invention comprise at least one dispatcher, at least one processing unit, at least one program memory, at least one program address generator, at least one data memory. Instruction decoding is performed in two stages. First the dispatcher decodes a category from each instruction and dispatches instructions to processing units that decode the remaining processing unit specific portion of the instruction to complete the execution.

The invention discloses a quantum dynamic computation method for lead compounds. The quantum dynamic computation method for the lead compounds includes the steps that a quantum dynamic model base of the lead compounds is established; related information of the lead compounds is quantized, and a database of the lead compounds is established; active compounds are discovered at random through quantumscreening; the lead compounds and active ingredients meeting required properties are searched in a database system, true random numbers generated through a quantum dynamic model are combined, and various compound models are established through a quantum screening mechanism; through the quantum dynamic model, mutual effects between active compound molecules and mutual effects between biological cells are established. The chemical effect is checked according to a quantum dynamic evolution mechanism; a synthetic active compound is subjected to a later quantum test simulation experiment, and a quantum dynamic model when compound molecules are combined with target spot protein is established. According to the quantum dynamic computation method for the lead compounds, the mutual effects betweenmolecular structures are subjected to evolution analysis with the quantum dynamic method, and the lead compounds corresponding to some pathologic structure can be obtained at higher accuracy and lower cost.

The invention provides a digital inner product calculator which is used for realizing various digital sequence calculations by taking digital inner products as a basic calculation form. The calculator comprises a statistical module and a first moment calculation module, wherein the statistical module is used for converting an inner product form into a first moment form through value statistics; the first moment calculation module is used for rapidly calculating the first moment acquired by the statistical module to obtain an inner product calculation result; and the digital inner product calculator comprises a coordination module which is used for connecting the statistical module and the first moment calculation module and realizing the coordination operation of the two modules. In the calculator, only a summator and a register are adopted; and when the inner product form changes, the calculation of inner products in different forms can be realized by changing the data input mode, therefore, the structure of the calculator per se is completely unnecessary to be changed.

The invention discloses a cloud-computing-based large scale curved surface reconstruction system. The system comprises a cloud computing platform deployed with a Hadoop framework. The cloud computing platform includes a main node and a sub-node. The main node is provided with a Hadoop user terminal, a NameNode element, and a JobTracker element. The sub-node is provided with a DataNode element and a TaskTracker element. According to the invention, through the introduction of cloud computing into a curved surface reconstruction system and through the use of the advantages of cloud computing such as high calculation capability and high system parallelism, the curved surface reconstruction capability of a curved surface reconstruction system is increased greatly. The reconstruction system has high expandability, utilizes resources properly and achieves low coupling. With the system, the problems with the insufficiency in single machine's memory and computing ability and the topological merging of the curved surfaces are solved.