80 results about "Process (computing)" patented technology

The invention discloses a method for partitioning dynamic tasks of CPU and GPU based on load balance, aiming at solving the technical problem of realizing the load balance between CPU and GPU and among multiple cores of CPU. The technical scheme of the method is as follows: establishing two databases for the task partitioning between CPU and GPU and among multiple cores of CPU; recording the partitioning proportion of tasks, wherein the partitioning proportion of tasks between CPU and GPU is determined according to problem scales; automatically adjusting the partitioning proportion of tasks in the whole executing processes of an application according to the actual measurement performance of CPU and GPU, thus realizing the load balance of task partitioning. The adoption of the method can ensure the load balance of tasks of CPU and GPU, improve the performance that a scientific computing program utilizes the GPU to accelerate, balance the task load among all the cores of CPU, and improve the performance of the multi-core CPU.

The invention provides a load balancing optimization method based on CPU (central processing unit) and MIC (many integrated core) framework processor cooperative computing, and relates to an optimization method for load balancing among computing nodes, between CPU and MIC computing equipment in the computing nodes and among computing cores inside the CPU and MIC computing equipment. The load balancing optimization method specifically includes task partitioning optimization, progress/thread scheduling optimization, thread affinity optimization and the like. The load balancing optimization method is applicable to software optimization of CPU and MIC framework processor cooperative computing, software developers are guided to perform load balancing optimization and modification on software of existing CPU and MIC framework processor cooperative computing modes scientifically, effectively and systematically, maximization of computing resource unitization by the software is realized, and computing efficiency and overall performance of the software are remarkably improved.

Techniques for efficient program deobfuscation through system application program interface (API) instrumentation are disclosed. In some embodiments, a system/process/computer program product for efficient program deobfuscation through system API instrumentation includes monitoring changes in memory after a system call event during execution of a malware sample in a computing environment; and generating a signature based on an analysis of the monitored changes in memory after the system call event during execution of the malware sample in the computing environment.

Exploiting FPGAs for acceleration may be performed by transforming concurrent programs. One example mode of operation may provide one or more of creating synchronous hardware accelerators from concurrent asynchronous programs at software level, by obtaining input as software instructions describing concurrent behavior via a model of communicating sequential processes (CSP) of message exchange between concurrent processes performed via channels, mapping, on a computing device, each of the concurrent processes to synchronous dataflow primitives, comprising at least one of join, fork, merge, steer, variable, and arbiter, producing a clocked digital logic description for upload to one or more field programmable gate array (FPGA) devices, performing primitive remapping of the output design for throughput, clock rate and resource usage via retiming, and creating an annotated graph of the input software description for debugging of concurrent code for the field FPGA devices.

The invention particularly relates to a lightweight application adaptation method oriented to an edge computing heterogeneous environment. According to the lightweight application adaptation method oriented to the edge computing heterogeneous environment, an integrated development environment driven by an automatic assembly line is constructed by combining a cross compiling technology and a virtualization technology, compilers of different hardware environments are packaged, and different steps and tasks in the compiling process are automatically connected in series by using a containerized cross compiling environment, a unified workflow framework is provided for heterogeneous environments, codes needing to be compiled and a compiling tool are combined together by using a container persistence means, and a container automatically runs a compiling assembly line to finally generate an executable program of each heterogeneous environment. According to the lightweight application adaptation method oriented to the edge computing heterogeneous environment, an edge computing application across the heterogeneous environment can be conveniently constructed, various edge computing operating system environments are adapted, and a heterogeneous CPU environment is supported. The problem that applications need to be repeatedly compiled in a heterogeneous environment is solved.

A method of mapping processes to processors in a parallel computing environment where a parallel application is to be run on a cluster of nodes wherein at least one of the nodes has multiple processors sharing a common memory, the method comprising using compiler based communication analysis to map Message Passing Interface processes to processors on the nodes, whereby at least some more heavily communicating processes are mapped to processors within nodes. Other methods, apparatus, and computer readable media are also provided.

The invention discloses a heterogeneous program execution method which is suitable for being executed in a simulator, the simulator is suitable for residing in computing equipment, a target heterogeneous program is installed on the simulator, and the method comprises the following steps: when the simulator is started, obtaining the target heterogeneous program on the simulator; loading the target heterogeneous program to a memory address space of a simulator process; creating a virtual CPU according to the starting request of the target heterogeneous program; and executing the target heterogeneous program through the virtual CPU. The invention also discloses a corresponding device, computing equipment and a readable storage medium.

The invention discloses a process migration method, a computing device and a storage medium, and the process migration method comprises the following steps of calculating the predicted running time of each process in a processor according to the running information of the processes; according to the predicted running time of the processes, determining the process with the predicted running time smaller than a running time threshold value as a short running process; acquiring the running information of the processor, and determining whether the short running process is used as a to-be-migrated process according to the running information of the processor; if the short running process is the to-be-migrated process, determining a target processor for migrating the to-be-migrated process according to the running information of other processors; and migrating the to-be-migrated process to the target processor for operation. According to the present invention, the migration of all the short running processes can be avoided, the migration frequency of the short running processes is reduced, the problem of invalidation of cache and memory affinity caused by migration is reduced, and it is guaranteed that when the short running processes are migrated, the system performance can be improved, and the use efficiency of the processor is improved.

The invention provides a hot update method and device for Unity, computing equipment and a computer readable storage medium, and the method comprises the steps: adding an interpreter realized by C + + for an IL2CPP tool, and transforming a runtime only supporting AOT static compiling into a runtime supporting an AOT compiler and the interpreter; the metadata management module during operation is further transformed, and the function of dynamically loading the hot update program set dynamic library is achieved; in the calling process of the function, the function from the hot update program set points to the interpreter by intercepting the loading process of the program set, so that the interpretive execution of the function in the hot update program set is completed, the unification of function calling is realized, and the long-standing defects of the hot update scheme in the prior art are overcome.

The invention provides a method for managing a physical bare computer based on Ironic, and belongs to the field of cloud computing; after the bare computer is started through mini-ramdisk, the mini-ramdisk can access a complete memory space, then an ipa-boot-script process is started to complete downloading and starting work of an IPA, and finally management of the bare computer by the ironic conductor is achieved.

This technology relates to transferring state information between processes or active software programs in a computing environment where the a new instance of a process or software program may receive such state information even after an original or old instance of the process or software program that owned the state information has terminated either naturally or unnaturally.

The invention relates to a distributed cluster resource scheduling method based on a user operation process. The method comprises the following steps: firstly, analyzing an execution sequence constraint relationship among sub-jobs included in the job, and determining a serial-parallel execution sequence of the sub-jobs; and according to the execution sequence of the sub-jobs, performing logic allocation of resources for the sub-jobs and predicting the execution time of each sub-job under the resource allocation, and further predicting the execution time of the job by calculating a key path in the job process. Usually, the job submitted by the user has the constraint of the completion time, so that the predicted job completion time can be used as a basis whether the cluster can provide services for the user in time or not. Experiments prove that compared with a default resource allocation algorithm of Spark, the algorithm provided by the invention can shorten the job execution time by 16.81%. According to the algorithm provided by the invention, the degree of parallelism of sub-job operation can be improved, the job execution time is shortened, and the improvement of the service quality of mechanisms such as a cloud service platform, a supercomputing center and a data center is facilitated.

One or more embodiments of the invention provide a unit test method and apparatus, a computing device and a medium, so as to provide a finer-grained unit test case management framework, and under the condition that test codes in a kernel space of an operating system are loaded, the test codes in the kernel space of the operating system are loaded. At least one test module is generated for a test code in a user space of an operating system, each test module is used for managing at least one unit test case, and each unit test case is provided with a corresponding operation identifier, so that the test code can be tested based on the operation identifier corresponding to each unit test case. And executing the test code corresponding to the unit test case of which the operation identifier meets the target condition in the kernel space to obtain an operation log and an operation result of the executed test code, so that independent operation of one or several unit test cases is realized, a kernel module only containing the unit test case does not need to be rewritten, and the operation efficiency is improved. Therefore, the efficiency of the unit test process can be improved.

The invention belongs to the field of computers, and particularly discloses a computer test method. The method is applied to a test host and a tested system. A plurality of test programs and a daemonprocess for managing the test programs are installed in the tested system. The plurality of test programs are compiled under a unified test program framework, in the test process, the test host is incommunication connection with the daemon process, the test host sends a request to the daemon process, and the daemon process receives the request sent by the test host and starts, stops or finely adjusts the test programs on the tested system. Compared with the prior art, the technical scheme disclosed by the invention has the advantages that different types of load generation and fault injectionprograms are compiled by using a unified test program framework, faults and loads can be flexibly assembled, various computing loads can be generated as required, and software and hardware faults canbe flexibly injected.

The invention provides a program code monitoring method which comprises the steps that computer equipment executes a first code corresponding to a first program code in a virtual execution environment, the first code belongs to an external code, and the external code is a code called in the first program code except an internal code; the external code comprises a system code provided by an operating system of the computing equipment, and the internal code is a code of a process generated by the first program code; in the process of executing the first code, if the second code belongs to the internal code, before the second code is executed, the computer equipment switches the execution environment of the first program code into a simulation execution environment, and the second code is a code to be executed; the second code is executed in the simulated execution environment. According to the method, under the condition that the aim of monitoring the instruction level of the program code is achieved, the performance overhead is reduced, and the operation efficiency of the system is improved.

The invention discloses an FPGA-based design method for improving the speed of a BCPNN, and relates to the technical field of artificial intelligence, and the method comprises the steps: updating variables, weights and offsets of synaptic states in the BCPNN on hardware through modular design; and achieving exponential operation on the FPGA through the lookup table. through a parallel algorithm, the speed of the weight and bias updating process of the synaptic state in the BCPNN is increased; through module multiplexing of the adder and the multiplier, the resource overhead is reduced under the condition of keeping the same computing performance. The method provided by the invention not only has higher calculation performance, but also has higher calculation accuracy, and can effectively improve the weight and offset updating speed of the BCPNN.

The invention discloses an intrusion kernel detection method and device, computing equipment and a computer storage medium, and the method comprises the steps: extracting the feature information of anintrusion kernel process for an ROOT attack behavior, and determining a detection condition corresponding to the feature information, wherein the detection conditions comprise suspicious attack detection conditions and determined attack detection conditions; setting multi-stage detection at different module layers of the equipment to detect the to-be-detected process, wherein each stage of detection corresponds to different detection conditions; and determining whether the to-be-detected process is a process invading the kernel or not according to a detection result of the multi-stage detection. Corresponding detection conditions are determined according to feature information extracted from a process invading a kernel, multi-level detection is set in different module layers of equipmentto detect to-be-detected processes in a targeted mode, attack behaviors of the processes can be effectively detected before the to-be-detected processes attack ROOT or in the attack process, and therefore process execution is prevented, and the process from invading kernel is avoided.

Our machine architecture and machine procedures use robustness, unpredictability and variability to hinder malware infection. In some embodiments, our machine instruction opcodes are randomized. The computing behavior of our machine is structurally stable (invariant) to small changes made to its machine instructions. Our invention expands the engineering method of stability to a cryptographically stable machine that is resistant to malware sabotage by an adversary.

Our procedures use quantum randomness to build unpredictable stable instructions. Our machine procedures can execute just before running a program so that the computing task can be performed with a different representation of its instructions during each run. A process of hiding a key or data inside of random noise is described that protects the privacy of the machine instruction opcodes and operands. In some embodiments, quantum randomness generates random noise, using photonic emission with a light emitting diode.

The invention discloses an elastic process management system and method. According to the invention, concepts of tasks and resources are highly abstracted; all concrete tasks and resource processes are managed in a unified manner; unified management is carried out on concrete at an abstract level through a task manager and a resource manager; therefore, dynamic allocation of resources for each concrete task is realized, the problems of poor computer computing power and low computer performance caused by multiple processes and inconsistent consumed time between the processes are prevented, anddynamic allocation of computing resources for an application program is effectively realized.

The invention discloses a high-performance computing platform application packaging method and system and a storage medium, and the method comprises the steps: downloading a source code of application software, and compiling the source code into a binary file; writing an operation specification file of the application software; packaging the binary file and the operation specification file together to obtain packaged application software; and uploading the packaged application software to a central application server. According to the method, each application corresponds to one running specification file, through the corresponding running specification file, the application and the corresponding dependent program can be downloaded, and then the corresponding loading and configuration program is run, so that the software can be directly used in a downloading machine, the tedious process of compiling and installing is omitted, a large amount of time and energy are saved, and through the method, the operation efficiency is improved. According to the technical scheme, the software can directly run into the corresponding operating system and the corresponding hardware architecture, the original running efficiency of the software is reserved to the maximum extent under the condition that the tedious compiling process is reduced, and great convenience is also provided for software management.