-
公开(公告)号：US07689992B2
公开(公告)日：2010-03-30
申请号：US10877095
申请日：2004-06-25
IPC分类号： G06F9/46
CPC分类号： G06F9/526
摘要： Shared locks are employed for controlling a thread which extends across more than one protocol layer in a data processing system. The use of a counter is used as part of a data structure which makes it possible to implement shared locks across multiple layers. The use of shared locks avoids the processing overhead usually associated with lock acquisition and release. The thread which is controlled may be initiated in either an upper layer protocol or in a lower layer.
摘要翻译： 共享锁用于控制在数据处理系统中跨越多于一个协议层延伸的线程。 计数器的使用被用作数据结构的一部分，这使得可以跨多层实现共享锁。 共享锁的使用避免了通常与锁获取和释放相关的处理开销。 被控制的线程可以在上层协议或下层协议中启动。
-
2.发明授权公开(公告)号：US07522590B2
公开(公告)日：2009-04-21
申请号：US10685144
申请日：2003-10-14
IPC分类号： H04L12/28
CPC分类号： H04L29/06 , H04L69/164 , H04L69/326
摘要： Messages arriving at a receiver are managed to ensure proper ordering of the messages. To facilitate proper ordering, a message sequence number is used, as well as matching criteria to match a correctly sequenced message with a posted receive. In response to processing a message, a check is made as to whether previously out of order messages can now be processed.
摘要翻译： 到达接收机的消息被管理以确保消息的正确排序。 为了方便正确的排序，使用消息序列号，以及匹配标准以将正确排序的消息与发布的接收匹配。 响应于处理消息，检查是否可以处理先前不按顺序的消息。
-
公开(公告)号：US06799317B1
公开(公告)日：2004-09-28
申请号：US09604310
申请日：2000-06-27
IPC分类号： G06F300
CPC分类号： G06F9/544 , G06F9/4812
摘要： A method for transparently handling messages originating from local shared memory and from an external source. A device driver allows the local sender to identify and wake up a waiting receiver task thread, simulating a packet arrival hardware interrupt. Upon awakening, the receiver task thread examines both shared memory and hardware message queues. The method can use a software routine that simulates handling of an occurrence of a hardware interrupt. The method invokes a local notify system service module that passes a window number identifying a receiving task. The method invokes a wake thread module that passes awakens a thread associated with the window number, and examines the shared memory buffer for receipt of the local source message. The method then copies the local source message from the shared memory buffer to the receiving task.
摘要翻译： 一种透明地处理源自本地共享内存和来自外部源的消息的方法。 设备驱动程序允许本地发送方识别和唤醒等待的接收者任务线程，模拟数据包到达硬件中断。 唤醒后，接收者任务线程检查共享内存和硬件消息队列。 该方法可以使用模拟处理硬件中断发生的软件程序。 该方法调用传递标识接收任务的窗口号的本地通知系统服务模块。 该方法调用唤醒线程模块，该模块通过唤醒与窗口号相关联的线程，并检查共享内存缓冲区以接收本地源消息。 该方法然后将本地源消息从共享内存缓冲区复制到接收任务。
-
公开(公告)号：US07694310B2
公开(公告)日：2010-04-06
申请号：US11467946
申请日：2006-08-29
摘要： A method for implementing Message Passing Interface (MPI-2) one-sided communication by using Low-level Applications Programming Interface (LAPI) active messaging capabilities, including providing at least three data transfer types, one of which is used to send a message with a message header greater than one packet where Data Gather and Scatter Programs (DGSP) are placed as part of the message header; allowing a multi-packet header by using a LAPI data transfer type; sending the DGSP and data as one message; reading the DSGP with a header handler; registering the DSGP with the LAPI to allow the LAPI to scatter the data to one or more memory locations; defining two sets of counters, one counter set for keeping track of a state of a prospective communication partner, and another counter set for recording activities of local and Remote Memory Access (RMA) operations; comparing local and remote counts of completed RMA operations to complete synchronization mechanisms; and creating a mpci_wait_loop function.
摘要翻译： 一种通过使用低级应用程序编程接口（LAPI）活动消息传递功能实现消息传递接口（MPI-2）单向通信的方法，包括提供至少三种数据传输类型，其中一种用于发送消息 一个消息头大于一个数据包，数据收集和散布程序（DGSP）作为消息头的一部分放置; 通过使用LAPI数据传输类型允许多分组报头; 发送DGSP和数据作为一个消息; 用头处理程序读取DSGP; 将DSGP注册到LAPI以允许LAPI将数据分散到一个或多个存储器位置; 定义两组计数器，一个用于跟踪预期通信伙伴的状态的计数器集合，以及用于记录本地和远程存储器访问（RMA）操作的活动的另一个计数器集合; 比较完成的RMA操作的本地和远程计数以完成同步机制; 并创建一个mpci_wait_loop函数。
-
公开(公告)号：US06401216B1
公开(公告)日：2002-06-04
申请号：US09181981
申请日：1998-10-29
申请人： Kalman Zvi Meth , Anton Prenneis , Adnan M. Agbaria , Patrick Francis Caffrey , William Joseph Ferrante , Su-Hsuan Huang , Demetrios K. Michailaros , William Gole Tuel, Jr.
发明人： Kalman Zvi Meth , Anton Prenneis , Adnan M. Agbaria , Patrick Francis Caffrey , William Joseph Ferrante , Su-Hsuan Huang , Demetrios K. Michailaros , William Gole Tuel, Jr.
IPC分类号： G06F1100
CPC分类号： G06F11/1458
摘要： A checkpoint of a parallel program is taken in order to provide a consistent state of the program in the event the program is to be restarted. Each process of the parallel program is responsible for taking its own checkpoint, however, the timing of when the checkpoint is to be taken by each process is the responsibility of a coordinating process. During the checkpointing, various data is written to a checkpoint file. This data includes, for instance, in-transit message data, a data section, file offsets, signal state, executable information, stack contents and register contents. The checkpoint file can be stored either in local or global storage. When it is stored in global storage, migration of the program is facilitated. When a parallel program is to be restarted, each process of the program initiates its own restart. The restart logic restores the process to the state at which the checkpoint was taken.
摘要翻译： 采取并行程序的检查点，以便在程序重新启动的情况下提供程序的一致状态。 并行程序的每个过程都负责自己的检查点，但是每个进程检查点何时执行的时间是协调过程的责任。 在检查点期间，将各种数据写入检查点文件。 该数据包括例如转接消息数据，数据部分，文件偏移，信号状态，可执行信息，堆栈内容和寄存器内容。 检查点文件可以存储在本地或全局存储中。 当它存储在全局存储中时，便于程序的迁移。 当并行程序重新启动时，程序的每个进程都会自动重启。 重新启动逻辑将进程恢复到执行检查点的状态。
-
6.发明授权公开(公告)号：US06338147B1
公开(公告)日：2002-01-08
申请号：US09182555
申请日：1998-10-29
申请人： Kalman Zvi Meth , Anton Prenneis , Adnan M. Agbaria , Patrick Francis Caffrey , William Joseph Ferrante , Su-Hsuan Huang , Demetrios K. Michailaros , William Gole Tuel, Jr.
发明人： Kalman Zvi Meth , Anton Prenneis , Adnan M. Agbaria , Patrick Francis Caffrey , William Joseph Ferrante , Su-Hsuan Huang , Demetrios K. Michailaros , William Gole Tuel, Jr.
IPC分类号： G06F1100
CPC分类号： G06F11/1458
摘要： A checkpoint of a parallel program is taken in order to provide a consistent state of the program in the event the program is to be restarted. Each process of the parallel program is responsible for taking its own checkpoint, however, the timing of when the checkpoint is to be taken by each process is the responsibility of a coordinating process. During the checkpointing, various data is written to a checkpoint file. This data includes, for instance, in-transit message data, a data section, file offsets, signal state, executable information, stack contents and register contents. The checkpoint file can be stored either in local or global storage. When it is stored in global storage, migration of the program is facilitated. When a parallel program is to be restarted, each process of the program initiates its own restart. The restart logic restores the process to the state at which the checkpoint was taken.
摘要翻译： 采取并行程序的检查点，以便在程序重新启动的情况下提供程序的一致状态。 并行程序的每个过程都负责自己的检查点，但是每个进程检查点何时执行的时间是协调过程的责任。 在检查点期间，各种数据被写入检查点文件。 该数据包括例如转接消息数据，数据部分，文件偏移，信号状态，可执行信息，堆栈内容和寄存器内容。 检查点文件可以存储在本地或全局存储中。 当它存储在全局存储中时，便于程序的迁移。 当并行程序重新启动时，程序的每个进程都会自动重启。 重新启动逻辑将进程恢复到执行检查点的状态。
-
公开(公告)号：US20080127203A1
公开(公告)日：2008-05-29
申请号：US11467946
申请日：2006-08-29
IPC分类号： G06F3/00
摘要： A method for implementing Message Passing Interface (MPI-2) one-sided communication by using Low-level Applications Programming Interface (LAPI) active messaging capabilities, including providing at least three data transfer types, one of which is used to send a message with a message header greater than one packet where Data Gather and Scatter Programs (DGSP) are placed as part of the message header; allowing a multi-packet header by using a LAPI data transfer type; sending the DGSP and data as one message; reading the DSGP with a header handler; registering the DSGP with the LAPI to allow the LAPI to scatter the data to one or more memory locations; defining two sets of counters, one counter set for keeping track of a state of a prospective communication partner, and another counter set for recording activities of local and Remote Memory Access (RMA) operations; comparing local and remote counts of completed RMA operations to complete synchronization mechanisms; and creating a mpci_wait_loop function.
摘要翻译： 一种通过使用低级应用程序编程接口（LAPI）活动消息传递功能实现消息传递接口（MPI-2）单向通信的方法，包括提供至少三种数据传输类型，其中一种用于发送消息 一个消息头大于一个数据包，数据收集和散布程序（DGSP）作为消息头的一部分放置; 通过使用LAPI数据传输类型允许多分组报头; 发送DGSP和数据作为一个消息; 用头处理程序读取DSGP; 将DSGP注册到LAPI以允许LAPI将数据分散到一个或多个存储器位置; 定义两组计数器，一个用于跟踪预期通信伙伴的状态的计数器集合，以及用于记录本地和远程存储器访问（RMA）操作的活动的另一个计数器集合; 比较完成的RMA操作的本地和远程计数以完成同步机制; 并创建一个mpci_wait_loop函数。
-
公开(公告)号：US20050289550A1
公开(公告)日：2005-12-29
申请号：US10877095
申请日：2004-06-25
申请人： Robert Blackmore , Su-Hsuan Huang , Chulho Kim , Richard Treumann , Hanhong Xue
发明人： Robert Blackmore , Su-Hsuan Huang , Chulho Kim , Richard Treumann , Hanhong Xue
IPC分类号： G06F9/46
CPC分类号： G06F9/526
摘要： Shared locks are employed for controlling a thread which extends across more than one protocol layer in a data processing system. The use of a counter is used as part of a data structure which makes it possible to implement shared locks across multiple layers. The use of shared locks avoids the processing overhead usually associated with lock acquisition and release. The thread which is controlled may be initiated in either an upper layer protocol or in a lower layer.
摘要翻译： 共享锁用于控制在数据处理系统中跨越多于一个协议层延伸的线程。 计数器的使用被用作数据结构的一部分，这使得可以跨多层实现共享锁。 共享锁的使用避免了通常与锁获取和释放相关的处理开销。 被控制的线程可以在上层协议或下层协议中启动。
-
9.发明申请公开(公告)号：US20050078605A1
公开(公告)日：2005-04-14
申请号：US10685144
申请日：2003-10-14
申请人： Su-Hsuan Huang , William Tuel
发明人： Su-Hsuan Huang , William Tuel
CPC分类号： H04L29/06 , H04L69/164 , H04L69/326
摘要： Messages arriving at a receiver are managed to ensure proper ordering of the messages. To facilitate proper ordering, a message sequence number is used, as well as matching criteria to match a correctly sequenced message with a posted receive. In response to processing a message, a check is made as to whether previously out of order messages can now be processed.
摘要翻译： 到达接收机的消息被管理以确保消息的正确排序。 为了方便正确的排序，使用消息序列号，以及匹配标准以将正确排序的消息与发布的接收匹配。 响应于处理消息，检查是否可以处理先前不按顺序的消息。
-
10.发明授权Concurrent access of an unsegmented buffer by writers and readers of the buffer 失效由缓冲区的写入者和读者同时访问未分段的缓冲区公开(公告)号：US06658525B1
公开(公告)日：2003-12-02
申请号：US09672642
申请日：2000-09-28
IPC分类号： G06F1300
CPC分类号： G06F13/4059
摘要： Data is written to an unsegmented buffer located within shared memory. While data is being written to the unsegmented buffer, at least a portion of the data is being read from the buffer. A counter is used to indicate how much space is available in the buffer to receive data. Further, the counter is employed to ensure that the reader does not advance beyond the writer.
摘要翻译： 数据被写入位于共享存储器内的未分段缓冲器。 当数据被写入未分段的缓冲器时，数据的至少一部分正在从缓冲器读取。 计数器用于指示缓冲区中有多少空间可用于接收数据。 此外，计数器用于确保读者不超越作者。
-
-
-
-
-
-
-
-
-
搜索结果
11 条记录 (耗时 0.019 秒)
