Intel Corporation

Google Inc.

Ivast, Inc.

Mikael Bourges-Sevenier

Justin E Gottschlich

Sridhar G Sharma

Adam Herr

Derek Gerstmann

Todd A Anderson

Justin Gottschlich

Methods, apparatus, systems and articles of manufacture are disclosed for intentional programming for heterogeneous systems. An example apparatus includes a code lifter to identify annotated code corresponding to an algorithm to be executed on the heterogeneous system based on an identifier being associated with the annotated code, and convert the annotated code in the first representation to intermediate code in a second representation by identifying the intermediate code as having a first algorithmic intent that corresponds to a second algorithmic intent of the annotated code, a domain specific language (DSL) generator to translate the intermediate code in the second representation to DSL code in a third representation when the first algorithmic intent matches the second algorithmic intent, the third representation corresponding to a DSL representation, and a code replacer to invoke a compiler to generate an executable including variant binaries based on the DSL code.

Methods and apparatus for intentional programming for heterogeneous systems

Methods, apparatus, systems and articles of manufacture are disclosed for intentional programming for heterogeneous systems. An example non-transitory computer readable storage medium includes instructions that, when executed, cause processor circuitry to at least identify a first code block having a first algorithmic purpose based on a second code block having a second algorithmic purpose, the second algorithmic purpose corresponding to the first algorithmic purpose, translate the first code block into executable domain specific language code, and output the executable domain specific language code.

Methods, apparatus, systems and articles of manufacture to optimize execution of a machine learning model are disclosed. An example apparatus includes a quantizer to quantize a layer of a model based on an execution constraint, the layer of the model represented by a matrix. A packer is to pack the quantized layer of the matrix to create a packed layer represented by a packed matrix, the packed matrix having non-zero values of the matrix grouped together along at least one of a row or a column of the matrix. A blocker is to block the packed layer into a blocked layer by dividing the non-zero values in the packed matrix into blocks. A fuser is to fuse the blocked layer into a pipeline. A packager is to package the pipeline into a binary.

Methods and apparatus to optimize execution of a machine learning model

Methods, apparatus, systems and articles of manufacture are disclosed improve utilization of a heterogeneous system executing software. The disclosed methods, apparatus, systems and articles of manufacture include an apparatus comprising a variant manager to determine whether an algorithm is a candidate for sub-algorithmic partitioning (SAP) based on at least one of a first size of input data to the algorithm and a second size of output data from the algorithm; a partitioner to partition the algorithm into at least a first tile and a second tile; and a compiler to compile a first variant based on the first tile and a second variant based on the second tile into an executable file, the first variant to be executed on a first processing element of the heterogeneous system, the second variant to be executed on a second processing element of the heterogeneous system.

Methods and apparatus to improve utilization of a heterogeneous system executing software

Methods, apparatus, systems and articles of manufacture are disclosed for runtime scheduling of software executing on a heterogeneous system. An example apparatus includes in response to a variant compiler to generate a representation of an algorithm in a domain-specific language (DSL), a compilation auto-scheduler to generate a schedule based on configurations for processing elements of the heterogeneous system, the processing elements including at least a first and a second processing element, the variant compiler to compile variant binaries based on the schedule, each of the variant binaries associated with the algorithm in the DSL, the variant binaries including a first variant binary corresponding to the first processing element and a second variant binary corresponding to the second processing element, and an application compiler to generate a fat binary including a runtime scheduler to select one or more of the variant binaries to execute a workload based on the schedule.

Methods and apparatus for runtime multi-scheduling of software executing on a heterogeneous system

Disclosed are machine processors and methods performed thereby. The processor has access to processing units for performing data processing and to libraries. Functions in the libraries are implementable to perform parallel processing and graphics processing. The processor may be configured to acquire (e.g., to download from a web server) a download script, possibly with extensions specifying bindings to library functions. Running the script may cause the processor to create, for each processing unit, contexts in which functions may be run, and to run, on the processing units and within a respective context, a portion of the download script. Running the script may also cause the processor to create, for a processing unit, a memory object, transfer data into that memory object, and transfer data back to the processor in such a way that a memory address of the data in the memory object is not returned to the processor.

Provision of a download script

A processor has access to processing units for performing data processing and to libraries. Functions in the libraries are implementable to perform parallel processing and graphics processing. The processor may be configured to acquire (e.g., to download from a web server) a download script, possibly with extensions specifying bindings to library functions. Running the script may cause the processor to create, for each processing unit, contexts in which functions may be run, and to run, on the processing units and within a respective context, a portion of the download script. Running the script may also cause the processor to create, for a processing unit, a memory object, transfer data into that memory object, and transfer data back to the processor in such a way that a memory address of the data in the memory object is not returned to the processor.

System and method of processing buffers in an OpenCL environment

Machine processor

An apparatus and method of processing a data file, by determining a range of values assumed by parameters in a hierarchy of nodes. The processing includes calculating a quantization parameter, based on an acceptable distortion level, indicating a desired number of bits, to represent the value range of each parameter for each node. Each node is examined to determine if a change in quantization parameter for a node parameter is desired, and if so a quantization parameter is inserted into the node. Then the hierarchy is reviewed, examining the quantization parameter at each node of the hierarchy and determining if one or more quantization parameters at a child node of the node being examined can be subsumed under the examined node quantization parameter. Quantization parameters are also used to encode animation data. The animation data, or frames, may be encoded into a sequence of Intra frames and Predictive frames. An Intra frame establishes quantization parameters used for subsequent predictive frames.

Growing community of inventors

Santa Clara, CA, United States of America

Mikael Bourges-Sevenier