414 results about "Cpu load" patented technology

Methods and apparatuses to manage working states of a data processing system. At least one embodiment of the present invention includes a data processing system with one or more sensors (e.g., physical sensors such as tachometer and thermistors, and logical sensors such as CPU load) for fine grain control of one or more components (e.g., processor, fan, hard drive, optical drive) of the system for working conditions that balance various goals (e.g., user preferences, performance, power consumption, thermal constraints, acoustic noise). In one example, the clock frequency and core voltage for a processor are actively managed to balance performance and power consumption (heat generation) without a significant latency. In one example, the speed of a cooling fan is actively managed to balance cooling effort and noise (and/or power consumption).

The invention discloses an operation and maintenance automation system and method. The system comprises a data acquisition module, a pre-processing and storing module, a prediction module, an algorithm evaluation module and an operation and maintenance monitoring management module, wherein the data acquisition module is used for acquiring key performance indexes and running states of monitored units in an operation and maintenance system through a network management protocol or a log file; the pre-processing and storing module is used for performing pre-processing work and sorted storing on data acquired by the data acquisition module; the prediction module is used for performing predictions, including a CPU (Central Processing Unit) load prediction and a disk load prediction according to the data processed by the pre-processing and storing module; the algorithm evaluation module is used for establishing an evaluation criterion of a prediction algorithm and the prediction module, comparing an actual value with a predicted value of the prediction algorithm, and establishing a self-learning process; and the operation and maintenance monitoring management module is used for interacting with operation and maintenance management personnel. A load prediction mechanism and an algorithm prediction model are established in order to finish predictions specific to resource use situations of CPUs, memories, disks and the like. Alarm information is analyzed by further referring to a load prediction result in order to give a relevant auxiliary decision. Resource expansion and fault handling are realized in a way of using scripts, an API (Application Programming Interface) interface and the like.

Anomalous behavior in a distributed system is automatically detected. Metrics are gathered for transactions, subsystems and/or components of the subsystems. The metrics can identify response times, error counts and/or CPU loads, for instance. Baseline metrics and associated deviation ranges are automatically determined and can be periodically updated. Metrics from specific transactions are compared to the baseline metrics to determine if an anomaly has occurred. A drill down approach can be used so that metrics for a subsystem are not examined unless the metrics for an associated transaction indicate an anomaly. Further, metrics for a component, application which includes one or more components, or process which includes one or more applications, are not examined unless the metrics for an associated subsystem indicate an anomaly. Multiple subsystems can report the metrics to a central manager, which can correlate the metrics to transactions using transaction identifiers or other transaction context data.

The technology disclosed enables the automatic definition of monitoring alerts for a web page across a plurality of variables such as server response time, server CPU load, network bandwidth utilization, response time from a measured client, network latency, server memory utilization, and the number of simultaneous sessions, amongst others. This is accomplished through the combination of load or resource loading and performance snapshots, where performance correlations allow for the alignment of operating variables. Performance data such as response time for the objects retrieved, number of hits per second, number of timeouts per sec, and errors per second can be recorded and reported. This allows for the automated ranking of tens of thousands of web pages, with an analysis of the web page assets that affect performance, and the automatic alignment of performance alerts by resource participation.

Methods and apparatuses to manage working states of a data processing system. At least one embodiment of the present invention includes a data processing system with one or more sensors (e.g., physical sensors such as tachometer and thermistors, and logical sensors such as CPU load) for fine grain control of one or more components (e.g., processor, fan, hard drive, optical drive) of the system for working conditions that balance various goals (e.g., user preferences, performance, power consumption, thermal constraints, acoustic noise). In one example, the clock frequency and core voltage for a processor are actively managed to balance performance and power consumption (heat generation) without a significant latency. In one example, the speed of a cooling fan is actively managed to balance cooling effort and noise (and/or power consumption).

The present invention relates to an apparatus and method for throttling a clock of a bus used for data exchange between devices in a computer such as a portable computer or notebook. Methods according to the invention can set a throttle rate of a clock to a predetermined initial value, detect a current remaining battery capacity or a current load to the CPU, and adjust the set throttle rate to a prescribed or calculated value according to the detected remaining battery capacity or the CPU load. Thus, power consumption is reduced, and, in the case of a battery-powered computer, battery life and operating time are extended.

The invention is directed to managing an abnormal client termination of a network connection with a server. When a client establishes a connection with the server, the server provides various heartbeat values to the client. The server may also send heartbeat values at various other times based on a variety of factors. Heartbeat values may be based on a network load, a CPU load, or the like, as well as different states of the client. For example, one heartbeat value may be used when the client is in an idle state. Other heartbeat values may be used when the client is engaged in a VOIP session, in a videoconferencing session, a streaming video session, or the like. When the client changes state, a different heartbeat value may be used to automatically modify a frequency for sending the heartbeat signal to the server.

A method, system and computer-readable medium for scheduling tasks, wherein a task switch request is initially received. A scheduling processor prioritizes the available tasks and inserts a highest priority task state into a first address register associated with a CPU. Next, the CPU suspends operation of the currently executing task and inserts a state of the suspended task into a second address register associated with the CPU. The CPU loads the task state from the first address register associated with the CPU and resumes the loaded task loaded. The scheduling processor then retrieves the task state from the second address register by the scheduling processor and schedules the retrieved task for subsequent execution.

The invention provides a dispatching method of a virtual machine under a multi-core environment. The dispatching method divides zones for CUP resources according to dispatching strategy types when a system is started, monitors the CPU loading condition of all divided zones in real time when the system operates, and dynamically adjusts the size of the CPU resources in the divided zones. The dispatching method carries out dispatching in the same divided zone by using the same dispatching strategy, thus improving dispatching efficiency, achieving the purpose of resource loading balance by dynamically adjusting the size of the CPU resources of the divided zone, realizing the full utilization of resources and reducing waste of resources.

The invention relates to a dynamic voltage frequency adjustment method in the power management technology, which comprises steps: collecting different hardware signals; writing a hardware dynamic voltage frequency adjustment controller and a corresponding register of a CPU loading log; reading CPU loading log information via an operating system/software drive, and inputting the information into an interface module of the operating system/software drive after analyzing; and reading dynamic processing temperature compensation, the hardware dynamic voltage frequency adjustment controller, the information processed by the interface module of the operating system/software drive; outputting to a corresponding frequency adjustment and voltage adjustment unit via related algorithm analysis, so as to complete adjusting system real time dynamic voltage frequency. The invention adjusts working frequency according to system performance demands, and correspondingly reduces system working voltage and reduces dynamic power dissipation. The invention greatly improves shortcomings of short service life of the battery, bad system stability and reliability brought in by the traditional simple power management technology. The invention can use various embedded system devices such as the portable multimedia recording and playing device, and smart phone.

Provided is a multicast distribution system which can perform a multicast distribution by selecting a communication path satisfying a communication feature required by a multicast group. A distribution source router (Rs) performs a multicast distribution to routers (R1 to R5) constituting a multicast group based on an application via a mesh virtual network connected to a plurality of routers. When the CPU load ratio and the multicast group communication feature exceed threshold values, the distribution source router updates the number of branches and the depth of the multicast tree which distributes a packet from the router and starts is reconfiguration of the multicast tree. When reconfiguring the multicast tree, the distribution source router references a path feature of the mesh virtual network, selects a communication path having a sufficient path feature with a higher priority, and generates a routing table based on the multicast tree.

The invention relates to a hardware shunt method of an IP report. In the technical scheme of the invention, a received IP report is shunted by a piece of network card hardware; the source address and the target address of an IP report title are extracted when the IP report is received by a network card; Hash algorithm is then employed to calculate tread to which the IP report belongs; and a DMA channel scheduling module starts up a DMA engine according to tread number to transmit the report to a main storage buffer area to which the tread belongs; in order to support hardware shunt strategy of the IP report, an upper-layer software ensures each tread to have a special main storage buffer area; the network card starts up the threads of a plurality of IP packages to correspond to the treads of host processing IP packages one to one; the treads of the host processing IP packages directly acquires data from a memory buffer area for processing; therefore, the transmission of intermediate data does not need a CPU, thereby reducing the CPU load. The hardware tread number supported by the method is as high as 1024 and even as higher as 4096 or 8192. The best configuration is that each CPU is corresponding to one tread, so that each tread operates independently without mutual interference; moreover, the best performance is achieved due to the least system resource sharing.

It is an object to improve accuracy of estimation of CPU load by calculating the CPU load necessary for performing I/O emulation under the virtualized environment based on disk load and/or network load. In case of estimating CPU load of a server X which operates servers 10 to 12 as virtual servers, a CPU performance converting unit 223 obtains measured values of CPU load of the servers 10 to 12. A load converting unit 220 obtains an estimated value of the CPU load of the server X caused by I/Os of disks and/or network from disk load and/or network load of the servers 10 to 12. A CPU overhead calculating unit 224 obtains a coefficient showing CPU overhead caused by virtualization. A load estimating unit 225 estimates the CPU load of the server X using the above measured value, the above estimated value, and the above coefficient.

The invention discloses a multi-user computation unloading method and device based on a chemical reaction optimization algorithm. A computation unloading device collects real-time information of a mobile terminal and an edge server, and the real-time information comprises the CPU load, the current electric quantity, the network bandwidth and like information of the terminal and the state information of a virtual machine of the edge server, the completion time of an application and the energy consumption occupied weight of the terminal are adjusted so as to establish an application completion time model, a terminal energy consumption model and a scheduler module; according to the collected real-time information and the established model, a selection decision is worked out by adopting the multi-user computation unloading method based on the chemical reaction optimization algorithm to obtain a quasi-optimal result of an application program, thereby improving the application performance and reducing the terminal energy consumption.

An encryption information determination section in a communication device negotiates an encryption algorithm to be used for encrypted communications with an encryption information determination section in a network camera, which is a communication counterpart. At this time, the encryption algorithm to be selected is varied depending on the CPU load of the communication device. That is, if the CPU utilization rate is high, a low-load encryption algorithm is selected, and if the CPU utilization rate is low, a high-load encryption algorithm is selected. A encryption/decryption processing section performs code processing using the encryption algorithm selected by the encryption information determination section.

The invention discloses a control method and a control device of display refresh rate of a mobile terminal. The control method comprises the following steps: the first step is storing n range terminals defined in a load rate range between 0 and 100% of a central processing unit (CPU) of the mobile terminal, n display refresh rate values, and a correspondence table between the n range terminals and the n display refresh rate values, the second steps is detecting an average value of the load rate of the CPU within a preset time range when the mobile terminal runs, the third step is judging a range terminal to which the average value, detected in the second step, of the load rate of the CPU belongs, and the fourth step is obtaining a corresponding display refresh rate according to the range terminal to which the average value belongs in the third step and by inquiring the correspondence table in the first step, and using the corresponding display refresh rate as the refresh rate of a display of the mobile terminal. According to the control method and the control device of the display refresh rate of the mobile terminal, the display refresh rate of the display is adjusted according to the dynamic of the load rate of the CPU when the mobile terminal runs, using experience of a user is not affected and energy consumption of the system is reduced at the same time.

The invention discloses a method for accessing an access pint (AP) into an access controller (AC) in a local area network, which comprises the following steps that: the AC receives discovery request messages sent by any one AP and sends discovery response messages carrying the waiting time to the AP, wherein the waiting time is determined according to the self central processing unit (CPU) load of the AC and the association priority of the AP and is used for notifying the AP to randomly select a waiting time in the waiting time, and datagram transport layer security (DTLS) connection requests are sent when the randomly selected waiting time is reached after the discovery completion for carrying out the access. Based on the same invention idea, the invention also provides the AC and the AP. The problem of simulation correlation of the AC by a large number of APs can be avoided, and convenience is brought for the network maintenance.

The invention relates to the fields of cloud computing and network technique and in particular relates to a virtual machine failure detection and recovery method. The method comprises the following steps: 1, deploying an NFS network file system environment, establishing a mount point on an NFS server, putting a mirror image on the mount point of the NFS server, mounting shared storage on all the nodes, and establishing a virtual machine for the mirror image; 2, starting a detection node timer, detecting whether the network and the power supply are positioned in a normal state, otherwise transmitting a failure recovery instruction to a scheduler; 3, selecting available nodes by virtue of the scheduler according to the preset cpu load strategy and other scheduling rules; 4, regenerating a virtual machine which is the same as the original machine according to a configuration file generated in the process of establishing the virtual machine; and 5, starting the established virtual machine. According to the method disclosed by the invention, the problem that the virtual machine cannot be continuously used after physical node downtime is solved, and the method can be used for virtual machine failure detection and recovery.

A method of transferring entire frequency shadow owing to the pre-computed radiancey of GPU, (1) making use of the circumstance illuminance image to pursues the illumination to circumstance , getting the radiation delivery function B = TL, matrix T is radiation delivery matrix, L is an illuminant matrix; (2) pre-computing the radiation delivery matrix T; (3)getting sparse radiation delivery matrix when the pre-computed radiation delivery matrix T compressed in quantization owe to small echo alternation; (4) rearranging the sparse radiation delivery matrix in (3) to put the important matrix in the front part; (5) doing small echo alternation rapidly for L, getting sparse illuminant matrix which has been quantization compressed; (6) carrying out rapidly sparse matrix multiplication on T and L in GPU to accomplish the re illumination exaggeration. The invention make use of the data structure and algorithm according to the ability of GPU that computed in parallel, it can reach fairly good balance between CPU loads and GPU loads, exaggerating speed and exaggerating quality . It can reduce the use of memory and keep the quality of exaggerating at the same time, and the exaggerating speed has increased in wide-range. It has reached the purpose of exaggerating entire frequency shadow in real time.

The invention discloses a method and a device for balancing multi-core CPU load. The method comprises the following steps of acquiring utilization rate information and frequency information of each CPU core in the multi-core CPU; acquiring a first CPU core with the frequency satisfying the operation condition of a to-be-operated process and lowest utilization rate; configuring the to-be-operated process into the first CPU. By adopting the embodiment, after the load balance of multi-core CPU is carried out, the increment of the power consumption can be avoided.