Some embodiments are directed to a logging within a software application executed over an assembly of information processing devices. More particularly, some embodiments relate to a method allowing process logging in the case of a software application operating with several processes and/or threads.

An information processing apparatus, includes a memory; and a first processor coupled to the memory and configured to: identify a maximum operating frequency of each of a plurality of second processors, when executing a plurality of processes to be subjected to parallel processing by the plurality of second processors, measure a load value representing a magnitude of a load of each of the plurality of processes, and determine, based on the identified maximum operating frequency of each of the plurality of second processors and the measured load value of each of the plurality of processes, a specific processor as an assignment destination of each of the plurality of processes from the plurality of second processors.

An offloading server includes: a data transfer designation section configured to analyze reference relationships of variables used in loop statements in an application and designate, for data that can be transferred outside a loop, a data transfer using an explicit directive that explicitly specifies a data transfer outside the loop; a parallel processing designation section configured to identify loop statements in the application and specify a directive specifying application of parallel processing by an accelerator and perform compilation for each of the loop statements; and a parallel processing pattern creation section configured to exclude loop statements causing a compilation error from loop statements to be offloaded and create a plurality of parallel processing patterns each of which specifies whether to perform parallel processing for each of the loop statements not causing a compilation error.

A system includes a memory storing program versions in a program history and a processor in communication with the memory. The processor is configured to store program components of a version of a program in a first directory associated with the version and select first and second candidate versions of the program from the program history to compare to a test version of the program. The processor is also configured to compare the first and second candidate versions to the test version, estimate an amount of work to regenerate the program from each of the candidate versions, and select the candidate version associated with a lesser amount of work. Additionally, the processor is configured to regenerate the program using components from the selected candidate version to create a regenerated program, store components of the regenerated program in a second directory associated with test version, and test the regenerated program.

A system and method for the aggregation and grouping of previously identified, causally related abnormal operating condition, that are observed in a monitored environment, is disclosed. Agents are deployed to the monitored environment which capture data describing structural aspects of the monitored environment, as well as data describing activities performed on it, like the execution of distributed transactions. The data describing structural aspects is aggregated into a topology model which describes individual components of the monitored environments, their communication activities and resource dependencies and which also identifies and groups components that serve the same purpose, like e.g. processes executing the same code. Activity related monitoring data is constantly monitored to identify abnormal operating conditions. Data describing abnormal operating condition is analyzed in combination with topology data to identify networks of causally related abnormal operating conditions. Causally related abnormal operating conditions are then grouped using known topological resource and same purpose dependencies. Identified groups are analyzed to determine their root cause relevance.

An electronic device may receive information associated with another electronic device that is proximate to the electronic device. Then, the electronic device may identify the other electronic device based on an identifier (such as a Media Access Control address) that is included in the information that is associated with the other electronic device. Moreover, the electronic device may access a pattern of activity for a region that includes the electronic device, where the pattern of activity includes events and identifiers of one or more electronic devices in the region during the events. For example, the events may include reported criminal activity. Next, the electronic device may determine an association between the other electronic device and at least one of the events based on the pattern of activity. Furthermore, the electronic device may provide a notification based on the determined association.

Described herein are systems, methods, and software to enhance the management and deployment of data processing clusters in a computing environment. In one example, a management system may monitor data processing efficiency information for a cluster and determine when the efficiency meets efficiency criteria. When the efficiency criteria are met, the management system may identify a new configuration for the cluster and initiate an operation to implement the new configuration for the cluster.

A method for performance analysis of a software application, by its parallel execution in a cluster of reference servers, includes a first execution involving exchanges of useful data between computational and storage nodes of the cluster of servers executed by an interconnection network according to a predetermined protocol by encapsulating these useful data in messages of predetermined size; a second execution involving the same exchanges of useful data between the same computational and storage nodes of the cluster of servers executed by the same interconnection network according to the same protocol but with a different predetermined message size; an extrapolation of the software application performance comprising a simulation of a variation of a bandwidth of the interconnection network based on the difference in the predetermined size of the messages exchanged during the first and second executions.

A parallelization method includes: generating a profiling result by performing profiling on a target neural network based on model information of the target neural network and architecture information of a manycore system; determining an assignment strategy to assign a plurality of cores of each of a plurality of clusters of the manycore system to a plurality of layers of the target neural network, based on the profiling result; and generating a parallelization strategy for parallel processing of the manycore system based on the assignment strategy.

Described herein are systems, methods, and software to enhance the management and deployment of data processing clusters in a computing environment. In one example, a management system may monitor data processing efficiency information for a cluster and determine when the efficiency meets efficiency criteria. When the efficiency criteria are met, the management system may identify a new configuration for the cluster and initiate an operation to implement the new configuration for the cluster.