A compiler for a gate-based superconducting quantum computer compiles hybrid classical/quantum algorithms for quantum processing cells with different configurations. The compiler inputs the algorithm and outputs code in a target language executable by a quantum processing cell of a quantum processing system that can execute the algorithm. The compiler includes various functionality, such as: parsing, analyzing control flows, addressing, compressing, and translating. The compiler optimizes algorithms in various manners using the functionality. Some optimizations include addressing efficiently, compressing based on simulations, and translating for efficient execution of parametric functions. The compiler may function in the environment of a cloud quantum computing system. The cloud quantum computing system may receive algorithms from remote access nodes for execution on local classical and quantum computing systems.

A graphics processing unit and method for processing fragments in a graphics processing system which includes: (i) hidden surface removal logic configured to perform hidden surface removal on fragments, and (ii) processing logic configured to execute shader programs for fragments. Initial processing of fragments is performed at the hidden surface removal logic. Some of the fragments have a shader-dependent property. A shader program for a particular fragment having the shader-dependent property is split into two stages. The initial processing comprises performing a depth test on the particular fragment. In response to the particular fragment passing the depth test of the initial processing in the hidden surface removal logic, a first stage, but not a second stage, of the shader program is executed for the particular fragment at the processing logic. The first stage of the shader program has instructions for determining the property of the particular fragment.

Provided is an application optimization method and an electronic device supporting the same. According to an example embodiment, the application optimization method may include: determining whether a condition set with respect to a duration of an idle state of the electronic device is satisfied, selecting an application for which application optimization is to be performed based on an application usage record of a user of the electronic device in response to the set condition being satisfied, and generating an optimized application by performing the application optimization in the background for the selected application.

Apparatuses, systems, and techniques to combine operations. In at least one embodiment, a processor causes two or more dependent reduction operations to be combined into a software kernel.

A compile method for a neural network, the compile method includes receiving data related to the neural network, generating a grouped layer by grouping layers comprised in the neural network based on the data, generating a set of passes executable in parallel based on a dependency between a plurality of passes to process the neural network, generating a set of threads performing a plurality of optimization functions based on whether optimization operations performed by the optimization functions is performed independently for the layers, respectively, or sequentially based on a dependency between the layers, and performing compilation in parallel based on the grouped layer, the set of passes, and the set of threads.

A recording medium stores a program for generating a source code that indicates processing on a sparse matrix and for causing a computer to execute a process including: acquiring second codes by optimizing, with a convex polyhedral model, a first code in which loop processing on a matrix is written in a static control part format; converting the second codes into source code candidates, based on sparse matrix information that indicates a variable that represents a non-zero element of the sparse matrix, expression information that indicates an operation expression that corresponds to a function included in the second codes, and data type information that indicates a type to be used for the variable; and selecting the source code from among the source code candidates in accordance with evaluation of processing performance for the sparse matrix in a case where each of the source code candidates is used.

The technology disclosed herein enables a processor to receive program code comprising a plurality of program code instructions generated by a compiler in view of source code, identify, among the program code instructions, one or more optimizable instructions, wherein at least one of the optimizable instructions is associated with an execution characteristic, and the execution characteristic is associated with an optimization decision, identify a profiling instruction location associated with the at least one of the optimizable instructions, and add a profiling instruction to the program code at the profiling instruction location. The at least one profiling instruction comprises a profiling identifier, and causes the processing device to: generate a profiling information item in view of the execution characteristic of the optimizable instructions, and store the profiling information item in a persistent memory region at a memory location corresponding to the profiling identifier.

Apparatuses, systems, and techniques to automatically generate a reduced number of compute kernels for performing operations of one or more neural networks. In at least one embodiment, one or more operations of one or more neural network graph nodes of the one or more neural network are automatically adjusted to generate an optimized one or more operations that are compiled to generate the reduced number of compute kernels.

A method and system provide the ability to compile computer source code. The source code is pre-processed to generate pure source code that includes definitions required for interpretation. The pure source code is formalized in a compiler, into assembly language that is processor specific. The formalization includes determining a set of two or more optimization routines, randomly selecting a selected optimization routine from the set of two or more optimization routines, and applying the selected optimization routine to each segment of the pure source code in a serialized manner. An executable binary file is then output and executed based on the formalized pure source code.

According to some embodiments, a system and methods comprising receiving application code for an on-premise application at a custom code cloud enabler module, wherein the application code includes at least one package of a plurality of objects; providing a whitelist of a plurality of cloud elements for the plurality of objects; identifying a first enhancement point in a first application object of the plurality of objects, the first enhancement point including a first extension element; selecting a first cloud element from the whitelist of cloud elements; determining the selected first cloud element matches a structure definition of the first extension element; mapping one or more parameters of the first extension element to one or more parameters in the matched first cloud element generating a cloud code snippet for the first extension element based on the mapping; and executing the generated cloud code snippet for the first enhancement point as part of the cloud code on a cloud platform. Numerous other aspects are provided.