209 results about "Job queue" patented technology

A system is provided for processing a print job. A processing unit, such as a server, receives a plurality of print job files. Each print job file is associated with a data file. The print job files are maintained in a queue of print job files. The processing unit selects a print job file in the queue and processes a data structure indicating a plurality of transform processes and the availability of each indicated transform process to process a data file. The processing unit selects an available transform process, such as a RIP process, in response to processing the data structure and indicates in the data structure that the selected transform process is unavailable. The transform process processes the data file associated with the selected print job to generate a printer supported output data stream.

A method for scheduling jobs in a computer system includes storing a plurality of job queue files in a random access memory, accessing at least one of the job queue files stored in random access memory, and scheduling, based in part on the at least one job queue file, execution of a job associated with the at least one job queue file. In a more particular embodiment, a method for scheduling jobs in a computer system include storing a plurality of job queue files in a random access memory. The plurality of job queue files include information associated with at least one job. The method also includes storing the information external to the random access memory and accessing at least one of the job queue files stored in the random access memory. The method also includes scheduling, based at least in part on the at least one job queue file, execution of a job associated with at least one job queue file. In response to the scheduling, the method includes updating the information stored in the job queue file and random access memory and the information stored external to the random access memory.

The invention discloses a data parallel processing system based on Cassandra. The data parallel processing system based on the Cassandra comprises a Hadoop main node, a plurality of Hadoop auxiliary nodes and a Cassandra storage end arranged on the Hadoop auxiliary node, wherein the main node comprises a user interface module, a Cassandra inquiring module, a job scheduling module, a job queue module, and a job tracker, wherein the auxiliary node comprises a task tracker, an input module, an output module and a Mapreduce module, the user interface module is used for receiving a user request, and judging that the type of the user request is a data inquiring request, or a submitting data processing job request, or a job information inquiring request, if the type of the user request is the data inquiring request, the user interface module sends the data inquiring request to the Cassandra inquiring module, if the type of the user request is the submitting data processing job request or the job information inquiring request, and the user interface module sends the submitting data processing job request or the job information inquiring request to the job scheduling module. The data parallel processing system based on the Cassandra has the advantages of being high in reliability, good in expansibility, and high in a throughput rate. The data parallel processing system based on the Cassandra has the capacity of simply inquiring and rapidly responding to the data, and meanwhile has the complex processing capacity to mass data.

The invention discloses a multi-queue task scheduling method and related system and equipment, which are used for solving the problem that differential resource allocation cannot be performed for each job queue because the priority of the queue is not supported during the conventional multi-queue task scheduling. The multi-queue task scheduling method comprises the following steps that: a scheduling node judges whether a job queue with a resource utilization ratio less than a set utilization ratio threshold exists or not when a task acquisition request transmitted by a task execution node is received; if the job queue is available, the scheduling node schedules a task to the task execution node from the job queue with the resource utilization ratio less than the utilization ratio threshold; and if the job queue is not available, the scheduling node schedules the task to the task execution node from the job queue with the highest priority.

Multiple intelligent pickers for an automated data storage library, the library having a library controller which identifies the start and end locations of the received move jobs, and places the move jobs in a job queue. The picker processor receives information over a communication link from other pickers describing movement information for the current move job of each of the other pickers. Upon completion of a move job, the picker communicates with the library controller, selecting a move job which avoids interference with the movement of the other pickers. The movement for the selected move job is determined so as to avoid interference with the stored movement profiles of the other pickers. A movement profile may be communicated to the other pickers over the communication link, and the move job is conducted according to the movement profile. Alternatively, movement information comprising the current location and vector are communicated.

The invention relates to a multi-tenant resource optimization scheduling method facing different types of loads. The multi-tenant resource optimization scheduling method comprises the following steps: 1, submitting jobs by system tenants and adding the jobs into a job queue; 2, collecting job load information and sending the job load information to a resource manager; 3, judging different load types of the jobs according to the job load information by the resource manager, and sending type information to a job scheduler; 4, carrying out job scheduling according to the different load types by the job scheduler; if the job is a calculation intensive type job, scheduling at the node; if the job is an I / O intensive type job, delaying and waiting; and 5, collecting job scheduling decision-making information to a scheduling reconstruction decision-making device, reconstructing a target calculation node, and carrying out the job scheduling according to a final decision-making result. According to the method provided by the invention, a multi-tenant shared cluster is realized, so that the cost of establishing an independent cluster is reduced, and meanwhile, a plurality of tenants can share more big data set resources. The better data locality is realized facing optimization of the different types of loads, and the balance between the equity and efficiency in a job scheduling process can be realized very well; and calculation performances of the whole cluster, such as throughput rate and job responding time, are improved.

A method and apparatus are provided for scheduling tasks within a computing device such as a communication switch. When a task is to be scheduled, other tasks in the work queue are analyzed to see if any can be executed simultaneously with the task to be scheduled. If so, the two tasks are combined to form a combined task, and the combined task is placed within the job queue. In addition, if the computing device has insufficient resources to execute the task to be scheduled, the task is placed back into the work queue for future scheduling. This is done in a way which avoids immediate reselection of the task for scheduling. Task processing efficiency is increased, since combining tasks reduces the waiting time for lower priority tasks, and tasks for which there are insufficient resources are delayed only a short while before a new scheduling attempt, rather than rejecting the task altogether.

A method and apparatus for controlling access to solid state memory devices which may allow maximum parallelism on accessing solid state memory devices with minimal interventions from firmware. To reduce the waste of host time, multiple flash memory devices may be connected to each channel. A job / descriptor architecture may be used to increase parallelism by allowing each memory device to operate separately. A job may be used to represent a read, write or erase operation. When firmware wants to assign a job to a device, it may issue a descriptor, which may contain information about the target channel, the target device, the type of operation, etc. The firmware may provide descriptors without waiting for a response from a memory device, and several jobs may be issued continuously to form a job queue. After the firmware finishes programming descriptors, a sequencer may handle the remaining work so that the firmware may concentrate on other tasks.

A symbolic execution task is dynamically divided among multiple computing nodes. Each of the multiple computing nodes explores a different portion of a same symbolic execution tree independently of other computing nodes. Workload status updates are received from the multiple computing nodes. A workload status update includes a length of a job queue of a computing node. A list of the multiple computing nodes ordered based on the computing nodes' job queue lengths is generated. A determination is made regarding whether a first computing node in the list is underloaded. A determination is made regarding whether a last computing node in the list is overloaded. Responsive to the first computing node being underloaded and the last computing node being overloaded, a job transfer request is generated that instructs the last computing node to transfer a set of one or more jobs to the first computing node.

The present invention provides a method, program product, and information processing system that efficiently dispatches jobs from a job queue. The jobs are dispatched to the computational nodes in the system. First, for each job, the number of nodes required to perform the job and the required computational resources for each of these nodes are determined. Then, for each node required, a node is selected to determine whether a job scheduler has a record indicating if this node meets the required computational resource requirement. If no record exists, the job scheduler analyzes whether the node meets the computational resource requirements given that other jobs may be currently executing on that node. The result of this determination is recorded. If the node does meet the computational resource requirement, the node is assigned to the job. If the node does not meet the resource requirement, a next available node is selected. The method continues until all required nodes are assigned and the job is dispatched to the assigned nodes. Alternatively, if the number of required nodes is not available, it is indicated the job can not be run at this time.

A method and apparatus are provided for scheduling tasks within a computing device such as a communication switch. When a task is to be scheduled, other tasks in the work queue are analyzed to see if any can be executed simultaneously with the task to be scheduled. If so, the two tasks are combined to form a combined task, and the combined task is placed within the job queue. In addition, if the computing device has insufficient resources to execute the task to be scheduled, the task is placed back into the work queue for future scheduling. This is done in a way which avoids immediate reselection of the task for scheduling. Task processing efficiency is increased, since combining tasks reduces the waiting time for lower priority tasks, and tasks for which there are insufficient resources are delayed only a short while before a new scheduling attempt, rather than rejecting the task altogether.

Informing the user when consumables and / or customer replaceable units will have to be resupplied, changed and / or replaced, relative to the jobs sent to the image forming system, is a beneficial way to avoid unnecessary printing delays. When a job is added to a job queue or repositioned within the job queue, a warning can be associated with that job, and displayed to the user. With this information, the user can pre-emptively add a consumable and / or to replace a customer replaceable unit to avoid delays and maintain productivity. Alternatively, if such a system fault would occur prior to an urgent job reaching the top of the job queue, the user can manipulate the order and / or presence of the various jobs in the queue to ensure the high-priority job is completed before the consumable is fully exhausted and / or the customer replaceable unit reaches the end of its useful life.