Adding and removing processes in a single view
    1.
    发明授权
    Adding and removing processes in a single view 有权
    在单个视图中添加和删除进程

    公开(公告)号:US07107313B2

    公开(公告)日:2006-09-12

    申请号:US09833573

    申请日:2001-04-13

    Inventor: Roger A. Fleming

    Abstract: The present invention provides a system and method for promoting a probationary member to a full member in a distributed system. A probationary member includes a process in a distributed system that is intended to be added or intended to replace an existing process in the system. Prior to adding or replacing the probationary member, the probationary member is evaluated based on criteria associated with context information related to the system. Based on the evaluation, the probationary member may be promoted to a full member of the system or eliminated from the system. Therefore, the present invention provides the opportunity to evaluate the probationary member in the system, prior to the probationary member being promoted to a full member of the system. Additionally, the probationary member may be added or replace an existing process in a single view change. Therefore, fault tolerance is maintained when the probationary member is replacing a mirror in a fault-tolerant distributed system.

    Abstract translation: 本发明提供了一种用于在分布式系统中促成试用成员到完整成员的系统和方法。 试用成员包括分布式系统中旨在添加或旨在替换系统中现有进程的进程。 在添加或替换试用成员之前,根据与系统相关的上下文信息相关的标准评估试用成员。 根据评估,试用成员可以升级为系统的正式成员或从系统中删除。 因此,本发明提供了在试用成员晋升为系统正式成员之前,对系统中的试用成员进行评估的机会。 此外,试用成员可以在单个视图更改中添加或替换现有进程。 因此,当试用成员正在更换容错分布式系统中的镜像时,会维持容错。

    System and method for detecting process and network failures in a distributed system
    2.
    发明授权
    System and method for detecting process and network failures in a distributed system 有权
    分布式系统中检测过程和网络故障的系统和方法

    公开(公告)号:US06820221B2

    公开(公告)日:2004-11-16

    申请号:US09833650

    申请日:2001-04-13

    Inventor: Roger A. Fleming

    CPC classification number: G06F11/0757 G06F11/0715

    Abstract: The present invention provides a system and method of detecting a process failure and a network failure in a distributed system. The distributed system includes a plurality of processes, each executing on a host, operable to transmit messages (i.e., heartbeats) to each other on a network. A process in the system is operable to execute a process failure algorithm for detecting failure of a process in the system. The process failure algorithm includes calculating a difference in the period of time to receive a heartbeat from a first processes and a period of time to receive a heartbeat from a second process in the system. If the difference exceeds a process failure threshold, the second process is suspected of failing. A process in the system is also operable to execute a network failure algorithm for detecting failure of a network connecting a plurality of hosts in the system. The network failure algorithm includes detecting receipt of a heartbeat from any one of a plurality of processes in the system within a network failure time limit. If a heartbeat is not received prior to the expiration of the network failure time limit, the network in the system is suspected of failing.

    Abstract translation: 本发明提供一种检测分布式系统中的过程故障和网络故障的系统和方法。 分布式系统包括多个进程,每个进程在主机上执行,可操作以在网络上彼此传送消息(即心跳)。 系统中的过程可操作以执行用于检测系统中的过程失败的过程失败算法。 过程失败算法包括计算从第一进程接收心跳的时间段和从系统中的第二进程接收心跳的时间段的差异。 如果差异超过过程失败阈值,则怀疑第二个进程失败。 系统中的过程也可操作用于执行网络故障算法,用于检测连接系统中多个主机的网络的故障。 网络故障算法包括在网络故障时间限制内检测系统中多个进程中的任何一个进程的心跳的接收。 如果在网络故障时间限制到期之前没有收到心跳,则怀疑系统中的网络发生故障。

    System and method for detecting process and network failures in a distributed system having multiple independent networks
    3.
    发明授权
    System and method for detecting process and network failures in a distributed system having multiple independent networks 有权
    用于检测具有多个独立网络的分布式系统中的过程和网络故障的系统和方法

    公开(公告)号:US06782489B2

    公开(公告)日:2004-08-24

    申请号:US09833771

    申请日:2001-04-13

    Inventor: Roger A. Fleming

    CPC classification number: H04L41/0681 H04L43/0823 H04L43/10 H04L43/16

    Abstract: The present invention provides a system and method of detecting a process failure and a network failure in a distributed system. The distributed system includes at least two processes, each executing on a host, operable to transmit messages (i.e., heartbeats) to each other on a plurality of networks in the distributed system. A process in the system is operable to execute a network failure algorithm for detecting failure of a network in the system. The process failure algorithm includes calculating a difference in the period of time to receive a heartbeat on a first network from a process and a period of time to receive a heartbeat on a second network from the process. If the difference exceeds a network failure threshold, the second network is suspected of failing. A process in the system is also operable to execute a process failure algorithm. The process failure algorithm includes detecting receipt of a heartbeat from a process on any one of a plurality of networks in the system within a network failure time limit. If a heartbeat is not received on any of the networks, the process is suspected of failing.

    Abstract translation: 本发明提供一种检测分布式系统中的过程故障和网络故障的系统和方法。 分布式系统包括至少两个进程,每个进程在主机上执行,可操作以在分布式系统中的多个网络上彼此发送消息(即心跳)。 系统中的过程可操作以执行用于检测系统中网络故障的网络故障算法。 过程失败算法包括计算从进程接收第一网络上的心跳的时间段的时间差,以及从进程接收第二网络上的心跳的时间段。 如果差异超过网络故障阈值,则怀疑第二个网络出现故障。 系统中的过程也可操作以执行过程失败算法。 过程失败算法包括在网络故障时间限制内检测系统中多个网络中的任何一个网络上的进程的心跳接收。 如果任何网络上没有收到心跳,该过程被怀疑是失败的。

    Adaptive heartbeats
    6.
    发明授权
    Adaptive heartbeats 有权
    自适应心跳

    公开(公告)号:US06782496B2

    公开(公告)日:2004-08-24

    申请号:US09833572

    申请日:2001-04-13

    Inventor: Roger A. Fleming

    CPC classification number: G06F11/0715 G06F11/0757 G06F11/3055 G06F11/3072

    Abstract: The present invention provides a system and method of adjusting a heartbeat timeout utilized for monitoring a process in a distributed system. The distributed system includes a plurality of processes monitoring one another by transmitting messages (i.e., heartbeats) indicative of a process being operational. A first process monitoring a second process is operable to receive one or more heartbeats from the second process in the distributed system. If the first process fails to receive a heartbeat from the second process prior to an expiration of the heartbeat timeout, the second process is suspected of failing. If the first process receives a heartbeat from the second process prior to the expiration of the heartbeat timeout, the first process recalculates the heartbeat timeout. Recalculating the heartbeat timeout includes gradually increasing or decreasing the heartbeat timeout based on a period of time to receive a heartbeat. Thus, the heartbeat timeout is gradually increased or decreased based on observed conditions, and transient conditions, which may affect transmission within the distributed system, may be accounted for prior to suspecting a process failure.

    Abstract translation: 本发明提供一种调整用于监视分布式系统中的过程的心跳超时的系统和方法。 分布式系统包括通过发送指示正在操作的进程的消息(即心跳)来监视彼此的多个进程。 监视第二进程的第一进程可操作以从分布式系统中的第二进程接收一个或多个心跳。 如果第一个进程在心跳超时到期之前未能从第二个进程接收到心跳,则怀疑第二个进程失败。 如果第一个进程在心跳超时到期之前从第二个进程接收到心跳,则第一个进程重新计算心跳超时。 重新计算心跳超时包括基于接收心跳的时间段逐渐增加或减少心跳超时。 因此,基于观察到的条件,心跳超时逐渐增加或减少,并且可能在怀疑过程失败之前考虑可能影响分布式系统内的传输的瞬态条件。

    Bead retainer
    7.
    发明授权
    Bead retainer 失效
    珠保持器

    公开(公告)号:US5060706A

    公开(公告)日:1991-10-29

    申请号:US668662

    申请日:1991-03-13

    CPC classification number: B60C15/028

    Abstract: A device (20) for retaining the beads of a tire against the flanges of a rim. The bead retainer device (20) is capable of being manufally deformable in the axial direction. The bead retainer device is also capable of being manually deformable in the circumferential direction such as to be collapsible.

    Abstract translation: 一种用于将轮胎的胎圈保持抵靠轮缘的凸缘的装置(20)。 胎圈保持装置(20)能够沿轴向方向可变形。 胎圈保持装置还能够在圆周方向上可手动地变形,以便可折叠。

Patent Agency Ranking