Methods for Packet Scheduling with Order in Software-Based Parallel Processing
    1.
    发明申请
    Methods for Packet Scheduling with Order in Software-Based Parallel Processing 有权
    基于软件的并行处理中的订单分组调度方法

    公开(公告)号:US20140189695A1

    公开(公告)日:2014-07-03

    申请号:US13730319

    申请日:2012-12-28

    CPC classification number: G06F9/46 G06F9/4881 G06F9/52 H04L12/5693 H04L47/50

    Abstract: A method for parallel processing implemented by a first core in a network unit, comprising locking an ingress queue if the ingress queue is not locked by another core, searching for an unlocked task queue from a first default subset of a plurality of task queues when the ingress queue is locked by another core, wherein the first subset is different from a second default subset of the plurality of task queues from which a second core begins a search for an unlocked task queue, and searching a remainder of the plurality of task queues for an unlocked task queue when all of the first default subset of task queues are locked and the ingress queue is locked.

    Abstract translation: 一种用于由网络单元中的第一核实现的用于并行处理的方法,包括:如果所述入口队列未被另一核心锁定,则锁定入口队列;当所述入口队列不被另一核心锁定时,从所述多个任务队列的第一默认子集中搜索未锁定任务队列 入口队列被另一核心锁定,其中第一子集与多个任务队列的第二默认子集不同,第二核心开始搜索未锁定的任务队列,并且搜索多个任务队列的其余部分 当任务队列的所有第一个缺省子集都被锁定并且入口队列被锁定时,该解锁的任务队列。

    Methods for task scheduling through locking and unlocking an ingress queue and a task queue
    2.
    发明授权
    Methods for task scheduling through locking and unlocking an ingress queue and a task queue 有权
    通过锁定和解锁入口队列和任务队列来进行任务调度的方法

    公开(公告)号:US09274826B2

    公开(公告)日:2016-03-01

    申请号:US13730319

    申请日:2012-12-28

    CPC classification number: G06F9/46 G06F9/4881 G06F9/52 H04L12/5693 H04L47/50

    Abstract: A method for parallel processing implemented by a first core in a network unit, comprising locking an ingress queue if the ingress queue is not locked by another core, searching for an unlocked task queue from a first default subset of a plurality of task queues when the ingress queue is locked by another core, wherein the first subset is different from a second default subset of the plurality of task queues from which a second core begins a search for an unlocked task queue, and searching a remainder of the plurality of task queues for an unlocked task queue when all of the first default subset of task queues are locked and the ingress queue is locked.

    Abstract translation: 一种用于由网络单元中的第一核实现的用于并行处理的方法,包括:如果所述入口队列未被另一核心锁定,则锁定入口队列;当所述入口队列不被另一核心锁定时,从所述多个任务队列的第一默认子集中搜索未锁定任务队列 入口队列被另一核心锁定,其中第一子集与多个任务队列的第二默认子集不同,第二核心开始搜索未锁定的任务队列,并且搜索多个任务队列的其余部分 当任务队列的所有第一个缺省子集都被锁定并且入口队列被锁定时,该解锁的任务队列。

    Methods for task scheduling through locking and unlocking an ingress queue and a task queue

    公开(公告)号:US09201683B2

    公开(公告)日:2015-12-01

    申请号:US13730319

    申请日:2012-12-28

    Abstract: A method for parallel processing implemented by a first core in a network unit, comprising locking an ingress queue if the ingress queue is not locked by another core, searching for an unlocked task queue from a first default subset of a plurality of task queues when the ingress queue is locked by another core, wherein the first subset is different from a second default subset of the plurality of task queues from which a second core begins a search for an unlocked task queue, and searching a remainder of the plurality of task queues for an unlocked task queue when all of the first default subset of task queues are locked and the ingress queue is locked.

Patent Agency Ranking