81 results about "Multilevel queue" patented technology

A system and method for managing multiple queues of non-persistent messages in a networked environment are disclosed. A particular embodiment includes: receiving a non-persistent message at a gateway process, the message including information indicative of a named queue; mapping, by use of a data processor, the named queue to a queue node by use of a consistent hash of the named queue; mapping the message to a queue process at the queue node; accessing, by use of the queue process, a list of subscriber gateways; and routing the message to each of the subscriber gateways in the list.

In one embodiment, a method, comprising receiving at a receive side of a physical link a request to suspend transmission of data from a queue within a transmit side of a first stage of queues and to suspend transmission via a path including the physical link, a portion of the first stage of queues, and a portion of a second stage of queues. The method includes sending, in response to the request, a flow control signal to a flow control module configured to schedule transmission of the data from the queue within the transmit side of the first stage of queues. The flow control signal is associated with a first control loop including the path and differing from a second control loop that excludes the first stage of queues.

In one embodiment, a method includes sending a first flow control signal to a first stage of transmit queues when a receive queue is in a congestion state. The method also includes sending a second flow control signal to a second stage of transmit queues different from the first stage of transmit queues when the receive queue is in the congestion state.

Methods and apparatus are provided for a enterprise-wide work flow system that may encompass multiple geographically separate sites. The sites may be either permanently or transiently linked. A single computer network may accommodate multiple work flow systems and a single work flow system may be distributed over multiple local area networks. The system maintains the paradigm of one global queue per service and provides for individual work flow systems to export services to one another in an enterprise.

Some embodiments provide a queue management system that efficiently and dynamically manages multiple queues that process traffic to and from multiple virtual machines (VMs) executing on a host. This system manages the queues by (1) breaking up the queues into different priority pools with the higher priority pools reserved for particular types of traffic or VM (e.g., traffic for VMs that need low latency), (2) dynamically adjusting the number of queues in each pool (i.e., dynamically adjusting the size of the pools), (3) dynamically reassigning a VM to a new queue based on one or more optimization criteria (e.g., criteria relating to the underutilization or overutilization of the queue).

The packet relay apparatus is provided. The packet relay apparatus includes a receiver that receives a packet; and a determiner that determines to drop the received packet without storing the received packet into a queue among the multi-stage queue. The determiner determines to drop the received packet at a latter stage, based on former-stage queue information representing a state of a queue at any former stage which the received packet belongs to and latter-stage queue information representing a state of a queue at the latter stage which the received packet belongs to.

Methods and systems for managing network traffic by multiple constraints are provided. A received network packet is assigned to multiple queues where each queue is associated with a different constraint. A network packet traverses the queues once it satisfies the constraint associated with that queue. Once the network packet has traversed all its assigned queues the network packet is forwarded to its destination. Also, network activity, associated with higher priority applications which are not managed, is detected. When such activity is detected, additional constraints are added to the network packet as it traverses its assigned queues.

The invention discloses a page hot degree based heterogeneous memory management method. The page hot degree based heterogeneous memory management method is characterized by including the steps of (1), as for memory pages of a heterogeneous memory, collecting total accessed read-write data, partial accessed read-write data and recent accessed read-write data in the memory pages, and storing the read-write data in a multi-level queue to form a page access record queue; (2), determining page hot degree values and hot degree values according to the partial read-write data and the recent read-write data, and sequentially reserving N pages with page hot degree values, exceeding a preset hot degree threshold value, as hot pages and N pages with page cold degree values, exceeding a preset cold degree threshold value, as cold pages; (3), selecting one optional cold page to be combined with each hot page, estimating energy-saving values, sequencing the energy-saving values in a descending sequence, selecting the cold and hot pages to be matched, and migrating the pages according to match results so as to reduce energy consumption. By the aid of the page hot degree based heterogeneous memory management method, high performance of a dynamic random access memory can be used fully, and overall performance of a heterogeneous memory system can be improved.

The invention discloses a method and a system for network address translation based on priority queue. The method comprises the steps that: firstly, a NAT service processing unit divides the received data massages needing to be NAT processed into multiple queues having different priorities, and sends the data massages having different priorities to corresponding queues of a CPU queue unit, wherein the data massages meet the NAP processing condition; secondly, a queue reading unit reads the data massages of the highest-priority queue in the CPU queue unit, the data massages is upward sent and network address translated, until the data massages of the highest-priority queue is empty, then the data massages of the next priority queue is read, upward sent and network address translated, the process is the same for other data massages. The invention improves the way of sending data massages upward, better guarantees the upward sending of high-priority data massages and generation of corresponding NAT converting item information, as well as application of high-priority users.

The invention discloses a method for realizing multilevel queue scheduling in a data network. The method comprises the following steps of: sending the message into the tail part of a corresponding queue according to the priority of the received message; if the exit direction of the current network is not congested, forwarding the message received currently; otherwise, according to the queue scheduling mode of strict priority queues, weighted fair queues and idle queues which are preset, firstly executing scheduling processing flow of the strict priority queues; then, if the exit direction of the current network also has residual bandwidth, executing scheduling processing flow of the weighted fair queues; and then, if the exit direction of the current network also has residual bandwidth, carrying out forward processing to the messages in the idle queues. The invention also provides a device used for realizing multilevel queue scheduling in the data network. The method can lead the messages without need of guarantee not to preempt the bandwidth, thus improving the flexible application capability of queue scheduling.

The invention relates to a user and service based QoS (quality of service) system in a Linux environment, which comprises a filter rule library, a filter, a classification rule library, a classifier and a queue maintaining module, wherein the filter rule library is connected with the filter; the classification rule library is connected with the classifier; the filter, the classifier and the queue maintaining module are successively connected; and the queue maintaining module adopts a structure of multistage queue plus LLQ (low latency queuing) and adopts a DWRR (display window right register) algorithm for dispatching and sending messages. Compared with the prior art, the system provided by the invention has the advantages of reducing sources required for system realization to the minimum, ensuring the system stability and the like.

The invention discloses a method and system for applying service to server from client-side. The keeping applying would be sent to server, when the applying from client-side is rejected by server, and the server would add the applying into queue until the condition being ready. The keeping condition would be set by user and periodically checking the validity of the applying, deleting the invalid applying. The waiting queue could be single stage or multi-stage, and the first applying of the highest priority would get the service first.

The invention discloses a dispatching management method for task queue priority of a large data set. According to the dispatching management method, a circulating thread for task acquisition is operated at each node of a task acquisition end, tasks are acquired from different queues according to priority of the queues, the tasks can be preferentially acquired from the high-priority queues, and meanwhile, the condition that the low-priority tasks cannot be blocked by the high-priority tasks is also taken into consideration. The queue service can support multiple queues of multiple types, the priority of each queue can be set, and average operating time of one queue can be acquired according to operating records; the number of the tasks acquired from each queue is dynamically adjusted according to multiple conditions including the task priority, queue length, the average operating time of each queue, the maximum number of the tasks acquired every time and the like, the tasks are dispatched reasonably, and resource waste is prevented.

The invention provides a queue scheduling method and system. The queue scheduling method comprises the following steps: gradually dividing a queue according to a preset standard and forming a tree-like multi-stage queue in which the number of sub-queues subordinate to the same parent queue is a preset value; caching data packets into interface layer queues corresponding to class mark numbers according to the class mark numbers carried in the received data packet; acquiring a queue marker of a scheduling queue according to scheduling information preset in a scheduling unit comprising the parent queue and all the sub-queues subordinate to the same parent queue, and carrying out the next-stage scheduling according to the scheduling information preset in the next-stage scheduling unit corresponding to the queue marker; and carrying out the above steps on each of the tree-like multi-stage queue from top to bottom until the scheduling queue is the interface layer queue. By the queue scheduling method and system in the invention, effective scheduling is carried out in a complicated multi-stage system, and quality of service (QOS) of a network is improved.

Technology for managing queuing resources of a shared network adapter is disclosed. The technology includes selectively transferring data from data transmission sources to a queue of the shared network adapter based on status indications from the shared network adapter regarding availability of queuing resources at the shared network adapter. In addition, the technology also includes features for selectively controlling transfer rates of data to the shared network adapter from applications, virtual network stations, other virtual adapters, or other data transmission sources. As one example, this selective control is based on how efficiently data from these data transmission sources are transmitted from the shared network adapter.

The embodiments of the present invention provide a method and device for controlling the allocation of a caching space and a computer storage medium. The method comprises: analyzing whether a queue to which a caching space to be shared is required to be allocated exists in each queue or not; when it is determined that the queue to which the caching space to be shared is required to be allocated exists, determining the number of queues to which the caching space to be shared is required to be allocated; and according to the determined number of the queues to which the caching space to be shared is required to be allocated, allocating the caching space to be shared to the queue to which the caching space to be shared is required to be allocated.

The invention discloses a server scheduling method and system. The method comprises the following steps: collecting resource consumption information of multiple servers; according to resource types, respectively establishing multiple server queues corresponding to different resource types, and adding the multiple servers into each server queue, wherein the servers in the server queues are ordered according to the resource consumption information of corresponding resource types; and receiving a server task instruction, according to the resource type consumed by the server task instruction, matching to the corresponding server queues, and according to a sequence, selecting servers in the server queues to execute the server task instruction. According to the invention, the servers with the best performance in a server cluster can be selected for executing the sever task instruction.

The invention discloses a replacement method and system for cached data in a storage system, and the storage system. The method comprises the steps of dividing a cache space of the storage system intoN cache modules in advance, and when the hit cache module exists in the N cache modules, accumulating a hit count of the hit cache module; inserting the hit cache module into an M-level queue with preset queue lengths of all levels, wherein the cache module hit for the first time is inserted in the first-level queue, and the queue of each level performs enqueuing operation and dequeuing operationaccording to an LRU replacement algorithm; and when the re-hit cache module exists in the ith-level queue, judging whether the cache module meets a condition of starting the (i+1)th-level queue or not according to a preset hit grading rule, and if yes, moving the cache module to the (i+1)th-level queue, or otherwise, performing queue adjustment on the ith-level queue according to the LRU replacement algorithm. The jitter problem of IO access performance is avoided; and the system access efficiency is improved.

The invention provides a multilevel menu page display method for an intelligent wearable device. A menu comprises multiple levels of pages which are displayed according to the levels and can execute device action instructions. The display method comprises the steps of storing and judging a current device state; and calling an action instruction callback function in a structure body, formed according to the multilevel menu, the corresponding pages of the multilevel menu and the action instructions, according to the current device state and a device action event, thereby enabling a device display module to perform recovery, maintenance or skip display of a page corresponding to a first-level menu in the multi-level menu. The invention furthermore discloses the intelligent wearable device. According to the method and the intelligent wearable device, the menu is divided through enumerated classification, so that the process is simplified and the multilevel menu is clearer; the skip or switching among different levels of pages is realized through a matching relationship between a structure body variable and the callback function; and the multilevel menu page display method for the intelligent wearable device, provided by the invention, has the advantages that the data structure is clearer, the control is easier, and the later addition and maintenance are facilitated.

The invention discloses a parallel task processing method based on task decomposition. According to the method, event trigger objects do not need to be sequentially transmitted by a plurality of queues, the queues into which the objects are inserted are decided by state parameters of the objects, threads do not need to know the task queues from which the objects operated by the threads are transmitted or the task queues from which the objects operated by the threads are about to be inserted, and the threads only need to complete self task processing. The event trigger objects, the task queues and the threads do not influence one another, and class packing is facilitated. When multiple subtasks of services have no order dependencies, the subtasks can be inserted into multiple queues at the same time, and parallelization processing of the subtasks is carried out. Processing thread sets of a fixed number are distributed to the tasks, and a dedicated thread does not need to be created for each object.

The invention discloses a multi-queue based scheduling method, which comprises the following steps: according to a multi-level stream classification strategy, performing circuit level mapping, user level mapping and/or service level mapping on a data packet, and putting the data packet in a mapping queue; and performing the scheduling in the circuit level, the user level and/or the service level according to a multi-queue scheduling algorithm. The invention also discloses a multi-queue based scheduling system, which comprises a multi-level strategy mapping module, a queue management module and a multi-level scheduling module, wherein the multi-level strategy mapping module is used for performing the circuit level mapping, the user level mapping and/or the service level mapping according to the multi-level stream classification strategy; the queue management module is connected with the multi-level strategy mapping module, and is used for putting the data packet in the mapping queue; and the multi-level scheduling module is connected with the queue management module, and is used for performing the scheduling in the circuit level, the user level and/or the service level according tothe multi-queue scheduling algorithm. Therefore, the method and the system can schedule data in the circuit level, the user level and the service level, realize complex traffic scheduling, and meet the increasingly complex QoS requirement.

Garbage collection methods include adding a data object to one of multiple queues owned by a first garbage collection thread. The queues include a public queue and multiple private queues. A task is popped from one of the plurality of queues to perform garbage collection. The public queue is swapped with one of the private plurality of private queues if there are no tasks in the public queue.