Data are maintained in a distributed computing system that describe a directed graph representing relationships among a set of items. The directed graph models a condition having an associated problem. The directed graph has graph components having associated data fields. The relationships are analyzed to identify a solution to the problem. As part of the analysis, a new value for the data field associated with a graph component is identified responsive to an operation performed during the analysis. The new value is compared with an existing value of the data field, and the data field is modified. The modification may include inserting the new value into an overflow vector of data, and replacing the existing value in the data field with exception information identifying the location of the new value. An exception flag associated with the data field is set to indicate that the exception information is being used.

Methods, apparatus and computer software product for source code optimization are provided. In an exemplary embodiment, a first custom computing apparatus is used to optimize the execution of source code on a second computing apparatus. In this embodiment, the first custom computing apparatus contains a memory, a storage medium and at least one processor with at least one multi-stage execution unit. The second computing apparatus contains at least one local memory unit that allows for data reuse opportunities. The first custom computing apparatus optimizes the code for reduced communication execution on the second computing apparatus. This Abstract is provided for the sole purpose of complying with the Abstract requirement rules. This Abstract is submitted with the explicit understanding that it will not be used to interpret or to limit the scope or the meaning of the claims.

A method of simulating codes that form a program configured to control a control target includes causing a processor to execute the following: first-setting a first process of executing a simulation of an operation of a code group excluding a specific code among the codes; second-setting a second process of executing a simulation of an operation of a specific model obtained by modeling the specific code; first-simulating, in the first process, a first simulation of a code preceding the specific code in the code group; second-simulating, in the second process, a second simulation of the specific model through use of an execution result of the first simulation by inter-process communication between the first process and the second process; and third-simulating, in the first process, a third simulation of a code succeeding the specific code through use of an execution result of the second simulation by the inter-process communication.

Data are received at a worker system in a distributed computing system that describe a graph representing relationships among a set of items. The graph models a condition having an associated problem. The graph has graph components having associated data fields. The received data are stored in a backup table, and the relationships are analyzed to identify a solution to the problem. As part of the analysis, a new value for the data field associated with a graph component is identified and compared with an existing value of the data field, and the data field is modified. The modified data field is stored in a delta table representing a change to the backup table.

A method for global data flow optimization for machine learning (ML) programs. The method includes receiving, by a storage device, an initial plan for an ML program. A processor builds a nested global data flow graph representation using the initial plan. Operator directed acyclic graphs (DAGs) are connected using crossblock operators according to inter-block data dependencies. The initial plan for the ML program is re-written resulting in an optimized plan for the ML program with respect to its global data flow properties. The re-writing includes re-writes of: configuration dataflow properties, operator selection and structural changes.

At the inputs and/or outputs, memories are assigned to a reconfigurable module to achieve decoupling of internal data and in particular decoupling of the reconfiguration cycles from the external data streams (to/from peripherals, memories, etc.)

This application provides a system architecture switching method and apparatus. The method includes: when a system architecture needs to be switched, transforming a first system architecture into a second system architecture, where the first system architecture represents a system architecture before switching; and providing a service for a user by using the second system architecture. Dynamic switching of a system architecture is implemented by using a transformable system architecture, so that switching of different architectures can be implemented by using only one system architecture. Therefore, only code for implementing the system architecture is required, and code overheads can be reduced in comparison with a conventional technology.

An offload server includes: an application code analysis section configured to analyze source code of an application; a data transfer designation section configured to, on the basis of a result of the code analysis, designate a data transfer to be collectively performed on, before starting GPU processing and after finishing the GPU processing, of variables that need to be transferred between a CPU and a GPU, those which are not mutually referenced nor mutually updated between CPU processing and the GPU processing and which are only to be returned to the CPU as a result of the GPU processing; a parallel processing designation section configured to identify loop statements in the application, and, for each of the identified loop statements, specify a statement specifying application of parallel processing by the GPU and perform compilation.

A method for coordinating the transfer of data between external memory and an array of data processors using address generators and local memory includes loading a plurality of groups of operands into local memory, processing the plurality of groups of operands on a single processor, and then returning the processed results to the external memory.

Data are received at a worker system in a distributed computing system that describe a graph representing relationships among a set of items. The graph models a condition having an associated problem. The graph has graph components having associated data fields. The received data are stored in a backup table, and the relationships are analyzed to identify a solution to the problem. As part of the analysis, a new value for the data field associated with a graph component is identified and compared with an existing value of the data field, and the data field is modified. The modified data field is stored in a delta table representing a change to the backup table.