54 results about "Workload scheduling" patented technology

A method of workload scheduling in a distributed compute environment includes assigning a subscriber of a network service to a first compute node instance (“CNI”) of a plurality of CNIs within a network node interposed between the subscriber and a provider of the network service. The subscriber traffic associated with the subscriber is processed at the first CNI. Subscriber specific data is generated at the first CNI related to the subscriber traffic. The subscriber specific data is then backed up to a second CNI of the network node that is designated as a standby CNI that will process the subscriber traffic if the first CNI fails.

Systems and methods implementing a demand-driven workload scheduling optimization of shared resources used to execute tasks submitted to a computer system are disclosed. Some embodiments include a method for demand-driven computer system resource optimization that includes receiving a request to execute a task (said request including the task's required execution time and resource requirements), selecting a prospective execution schedule meeting the required execution time and a computer system resource meeting the resource requirement, determining (in response to the request) a task execution price for using the computer system resource according to the prospective execution schedule, and scheduling the task to execute using the computer system resource according to the prospective execution schedule if the price is accepted. The price varies as a function of availability of the computer system resource at times corresponding to the prospective execution schedule, said availability being measured at the time the price is determined.

Various embodiments of methods and systems for thermally aware scheduling of workloads in a portable computing device that contains a heterogeneous, multi-processor system on a chip (“SoC”) are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit different processing efficiencies at a given temperature, and because more than one of the processing components may be capable of processing a given block of code, thermally aware workload scheduling techniques that compare performance curves of the individual processing components at their measured operating temperatures can be leveraged to optimize quality of service (“QoS”) by allocating workloads in real time, or near real time, to the processing components best positioned to efficiently process the block of code.

Method and system for providing a mechanism for determining an optimal workload distribution, from a plurality of candidate workload distributions, each of which has been determined to optimize a particular aspect of a workload-scheduling problem. More particularly, the preferred embodiment determines a workload distribution based on resource selection policies. From this workload distribution, the preferred embodiment optionally determines a workload distribution based on job priorities. From either or both of the above parameters, the preferred embodiment determines a workload distribution based on a total prioritized weight parameter. The preferred embodiment also determines a workload distribution which attempts to match the previously determined candidate workload distributions to a goal distribution. Similarly, the preferred embodiment calculates a further workload distribution which attempts to maximize job throughput.

Systems include one or more critical datacenter connected to behind-the-meter flexible datacenters. The critical datacenter is powered by grid power and not necessarily collocated with the flexible datacenters, which are powered “behind the meter.” When a computational operation to be performed at the critical datacenter is identified and determined that it can be performed more efficiently or advantageously at a flexible datacenter, the computational operation is instead obtained by the flexible datacenters for performance. The critical datacenter and flexible datacenters preferably share a dedicated communication pathway to enable high-bandwidth, low-latency, secure data transmissions. In some situations, a computational operation is supported by multiple datacenters in a redundant arrangement, such as multiple flexible datacenters.

The present invention relates to the field of enterprise network computing. In particular, it relates to a method and respective system for monitoring workload of a workload scheduler. Information defining a plurality of test jobs of low priority is received. The test jobs have respective launch times, and the test jobs are launched for execution in a data processing system in accordance with said launch times and said low execution priority. It is evaluated how many of said test jobs are executed within a pre-defined analysis time range. A performance decrease warning is issued, if the number of executed test jobs is lower than a predetermined threshold number. The workload scheduler discards launching of jobs having a low priority when estimating that a volume of jobs submitted with higher priority is sufficient to keep said scheduling system busy.

A method for scheduling execution of a work unit in a data processing system (100) comprises assigning to the work unit an expected execution duration (PD); executing the work unit (315); determining an actual execution duration (AD) of the work unit (320); determining a difference between the actual execution duration and the expected duration; and conditionally adjusting (345,350,355) the expected execution duration assigned to the work unit based on the measured actual execution duration, wherein the conditionally adjusting includes preventing the adjustment of the expected execution duration in case said difference exceeds a predetermined threshold (LF). The method further includes associating to the work unit a parameter (NLF) having a prescribed value adapted to provide an indication of unconditional adjustment of the expected execution duration: in case said parameter takes the prescribed value, the expected duration assigned with the work unit based on the measured actual execution duration even if the difference in durations exceeds the predetermined threshold.

The method of the present invention provides an automatic and optimised selection of the network topology for distributing scheduling of jobs on the computers of the modified network topology. The automatic and optimised selection of the network topology starts from the current topology and a desired number of additional connections. In this way the method of the present invention provides a higher convergence speed for the modified consensus algorithm in comparison, e.g., to a simple ring network. The method exploits the so called small-world networks. Small-world networks are more robust to perturbations than other network architectures. The preferred embodiment provides a workload scheduling system which is highly scalable to accommodate increasing workloads within a heterogeneous distributed computing environment. A modified average consensus algorithm is used to distribute network traffic and jobs amongst a plurality of computers.

Various embodiments of methods and systems for mode-based reallocation of workloads in a portable computing device (“PCD”) that contains a heterogeneous, multi-processor system on a chip (“SoC”) are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit different performance capabilities or strengths, and because more than one of the processing components may be capable of processing a given block of code, mode-based reallocation systems and methodologies can be leveraged to optimize quality of service (“QoS”) by allocating workloads in real time, or near real time, to the processing components most capable of processing the block of code in a manner that meets the performance goals of an operational mode. Operational modes may be determined by the recognition of one or more mode-decision conditions in the PCD.

A method of automatically optimizing workload scheduling. Target values for workload characteristics and constraint specifications are received. Generation of a first execution plan is initiated. Initial constraint values conforming to the constraint specifications are selected. Each constraint value constrains tasks included in the workload. The first execution plan is executed, thereby determining measurements of workload characteristics. Contributions indicating differences between workload characteristic measurements and target values are determined and stored. Generation of a next execution plan is initiated. Modified constraint values conforming to the constraint specifications are selected. Changes in the workload characteristics based on the modified constraint values are evaluated. An optimal or acceptable sub-optimal solution in a space of solutions defined by the constraint specifications is determined, resulting in new values for the constraints. After replacing the initial values with the new values, the next execution plan is generated and executed.

A method, system and computer program product for detecting workload imbalance in a dynamically scheduled computer server cluster are disclosed. One such method comprises the steps of: monitoring a plurality of metrics at each of said computer servers; detecting change points in said plurality of metrics; generating alert points based on said detected change points; alarm points are correlated; and based on a result of said correlation, identifying one or more of said computer servers causing a workload imbalance. Systems and computer program products for practicing the methods described above are also disclosed.

Operating system-directed workload scheduling of an adjunct partition in a logically partitioned computer is selectively overridden to handle platform work requiring a Quality of Service (QoS) guarantee. Firmware may track outstanding requests for platform work for an adjunct partition, and in response to a request for platform work that requires a QoS guarantee, the firmware may assume or take over scheduling decisions for the adjunct partition from the operating system of an associated logical partition and schedule execution of the adjunct partition to ensure that the adjunct partition will be allocated sufficient execution resources to perform the platform work independent of the scheduling desires of the operating system. As a result, any platform work that potentially impacts the platform work of other adjunct partitions will not be held up as a result of an unwillingness or inability of the operating system to schedule execution of the adjunct partition.

Systems, software, devices, and methods of distributing a workload among available data storage devices in a thermal aware manner are described herein. More specifically, the examples herein discuss distributing the workload among the available data storage devices in a thermal aware manner that optimizes collective IOPs of the data storage devices in an enclosure. The thermal aware distribution of the storage operations is determined by a thermal model that predicts thermal characteristics of the data storage system based on inlet air characteristics of the enclosure, performance characteristics and thermal constraints of the data storage devices, and constraints of the workload.

A method, system, and device for managing hardware resources in a cloud scheduling environment includes a zone controller. The zone controller can manage groups of node servers in a cloud datacenter using a checkin service. The checkin service allows server groups to be created automatically based on one or more hardware characteristics of the node servers, server health information, workload scheduling or facilities management parameters, and/or other criteria.

Various embodiments of methods and systems context-aware thermal management in a portable computing device (“PCD”) are disclosed. Notably, the environmental context to which a PCD is subjected may have significant impact on the PCD's thermal energy dissipation efficiency. Embodiments of the solution seek to leverage knowledge of a PCD's environmental context to modify or adjust thermal policy parameters applied within a PCD in response to a thermal event within the PCD.

The present application discloses a method and apparatus for assigning workload. The method includes the following steps that: to-be-assigned work information is obtained; the workload of the work information is calculated by using a predetermined workload algorithm; the current workload of assignable workers within a predetermined time period is obtained; and the workload of the work informationis assigned to a worker in the assignable workers who is capable of taking over the workload on the basis of the current workload, so that the ratio of the workload of the worker capable of taking over the workload to the workload of all workers in a team satisfies the preset work weight of the workers. With the method adopted, work can be automatically assigned according to the preset work weight, and workload scheduling can be recorded at the same time; the work can be assigned effectively; and the platformization and intellectualization of work scheduling can be realized.