专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08752051B2 Performing an allreduce operation using shared memory 失效
标题翻译：使用共享内存执行allreduce操作
公开(公告)号：US08752051B2
公开(公告)日：2014-06-10
申请号：US13427057
申请日：2012-03-22
申请人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46 , G06F9/48 , G06F9/52
CPC分类号： G06F9/4843 , G06F9/52 , G06F9/546
摘要： Methods, apparatus, and products are disclosed for performing an allreduce operation using shared memory that include: receiving, by at least one of a plurality of processing cores on a compute node, an instruction to perform an allreduce operation; establishing, by the core that received the instruction, a job status object for specifying a plurality of shared memory allreduce work units, the plurality of shared memory allreduce work units together performing the allreduce operation on the compute node; determining, by an available core on the compute node, a next shared memory allreduce work unit in the job status object; and performing, by that available core on the compute node, that next shared memory allreduce work unit.
摘要翻译：公开了用于使用共享存储器执行全部还原操作的方法，装置和产品，其包括：由计算节点上的多个处理核心中的至少一个接收执行全部降低操作的指令; 通过所述接收到所述指令的核心建立用于指定多个共享存储器全部还原工作单元的作业状态对象，所述多个共享存储器全部还原工作单元一起在所述计算节点上执行全部还原操作; 通过所述计算节点上的可用核确定所述作业状态对象中的下一个共享存储器allreduce工作单元; 并且通过计算节点上的可用核心执行下一个共享存储器allreduce工作单元。

2. 发明申请

US20080301683A1 Performing an Allreduce Operation Using Shared Memory 有权
标题翻译：使用共享内存执行Allreduce操作
公开(公告)号：US20080301683A1
公开(公告)日：2008-12-04
申请号：US11754782
申请日：2007-05-29
申请人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46
CPC分类号： G06F9/4843 , G06F9/52 , G06F9/546
摘要： Methods, apparatus, and products are disclosed for performing an allreduce operation using shared memory that include: receiving, by at least one of a plurality of processing cores on a compute node, an instruction to perform an allreduce operation; establishing, by the core that received the instruction, a job status object for specifying a plurality of shared memory allreduce work units, the plurality of shared memory allreduce work units together performing the allreduce operation on the compute node; determining, by an available core on the compute node, a next shared memory allreduce work unit in the job status object; and performing, by that available core on the compute node, that next shared memory allreduce work unit.
摘要翻译：公开了用于使用共享存储器执行全部还原操作的方法，装置和产品，其包括：由计算节点上的多个处理核心中的至少一个接收执行全部降低操作的指令; 通过所述接收到所述指令的核心建立用于指定多个共享存储器全部还原工作单元的作业状态对象，所述多个共享存储器全部还原工作单元一起在所述计算节点上执行全部还原操作; 通过所述计算节点上的可用核确定所述作业状态对象中的下一个共享存储器allreduce工作单元; 并且通过计算节点上的可用核心执行下一个共享存储器allreduce工作单元。

3. 发明申请

US20120179881A1 Performing An Allreduce Operation Using Shared Memory 失效
标题翻译：使用共享内存执行Allreduce操作
公开(公告)号：US20120179881A1
公开(公告)日：2012-07-12
申请号：US13427057
申请日：2012-03-22
申请人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F12/02
CPC分类号： G06F9/4843 , G06F9/52 , G06F9/546
摘要： Methods, apparatus, and products are disclosed for performing an allreduce operation using shared memory that include: receiving, by at least one of a plurality of processing cores on a compute node, an instruction to perform an allreduce operation; establishing, by the core that received the instruction, a job status object for specifying a plurality of shared memory allreduce work units, the plurality of shared memory allreduce work units together performing the allreduce operation on the compute node; determining, by an available core on the compute node, a next shared memory allreduce work unit in the job status object; and performing, by that available core on the compute node, that next shared memory allreduce work unit.
摘要翻译：公开了用于使用共享存储器执行全部还原操作的方法，装置和产品，其包括：由计算节点上的多个处理核心中的至少一个接收执行全部降低操作的指令; 通过所述接收到所述指令的核心建立用于指定多个共享存储器全部还原工作单元的作业状态对象，所述多个共享存储器全部还原工作单元一起在所述计算节点上执行全部还原操作; 通过所述计算节点上的可用核确定所述作业状态对象中的下一个共享存储器allreduce工作单元; 并且通过计算节点上的可用核心执行下一个共享存储器allreduce工作单元。

4. 发明授权

US08161480B2 Performing an allreduce operation using shared memory 有权
标题翻译：使用共享内存执行allreduce操作
公开(公告)号：US08161480B2
公开(公告)日：2012-04-17
申请号：US11754782
申请日：2007-05-29
申请人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Gabor Dozsa , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46 , G06F13/00 , G06F7/38
CPC分类号： G06F9/4843 , G06F9/52 , G06F9/546
摘要： Methods, apparatus, and products are disclosed for performing an allreduce operation using shared memory that include: receiving, by at least one of a plurality of processing cores on a compute node, an instruction to perform an allreduce operation; establishing, by the core that received the instruction, a job status object for specifying a plurality of shared memory allreduce work units, the plurality of shared memory allreduce work units together performing the allreduce operation on the compute node; determining, by an available core on the compute node, a next shared memory allreduce work unit in the job status object; and performing, by that available core on the compute node, that next shared memory allreduce work unit.
摘要翻译：公开了用于使用共享存储器执行全部还原操作的方法，装置和产品，其包括：由计算节点上的多个处理核心中的至少一个接收执行全部降低操作的指令; 通过所述接收到所述指令的核心建立用于指定多个共享存储器全部还原工作单元的作业状态对象，所述多个共享存储器全部还原工作单元一起在所述计算节点上执行全部还原操作; 通过所述计算节点上的可用核确定所述作业状态对象中的下一个共享存储器allreduce工作单元; 并且通过计算节点上的可用核心执行下一个共享存储器allreduce工作单元。

5. 发明授权

US08949453B2 Data communications in a parallel active messaging interface of a parallel computer 有权
标题翻译：并行计算机的并行活动消息接口中的数据通信
公开(公告)号：US08949453B2
公开(公告)日：2015-02-03
申请号：US12956903
申请日：2010-11-30
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46 , G06F9/54
CPC分类号： G06F9/546
摘要： Data communications in a parallel active messaging interface (‘PAMI’) of a parallel computer, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint; transmitting from the origin endpoint to the first target endpoint a Request-To-Send (‘RTS’) message advising the first target endpoint of the location and size of the transfer data; assigning by the first target endpoint to each of a plurality of target endpoints separate portions of the transfer data; and receiving by the plurality of target endpoints the transfer data.
摘要翻译：并行计算机的并行活动消息接口（“PAMI”）中的数据通信，由数据通信端点组成的PAMI，每个端点包括计算节点上的执行线程的数据通信参数的规范，包括客户端的规范，上下文和任务，通过PAMI和通过数据通信资源耦合用于数据通信的端点，包括在PAMI的原点端点接收SEND指令，SEND指令指定传输数据从原始端点传输到第一个目标终点; 从所述起始端点向所述第一目标端点发送向所述第一目标端点通知所述传送数据的位置和大小的请求发送（'RTS'）消息; 将第一目标端点分配给多个目标端点中的每一个分离传输数据的部分; 以及由所述多个目标端点接收所述传送数据。

6. 发明授权

US08910178B2 Performing a global barrier operation in a parallel computer 有权
标题翻译：在并行计算机中执行全局屏障操作
公开(公告)号：US08910178B2
公开(公告)日：2014-12-09
申请号：US13206581
申请日：2011-08-10
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46 , G06F9/52
CPC分类号： G06F9/522 , G06F9/46 , G06F9/52 , G06F9/542 , G06F9/546 , G06F15/17318
摘要： Executing computing tasks on a parallel computer that includes compute nodes coupled for data communications, where each compute node executes tasks, with one task on each compute node designated as a master task, including: for each task on each compute node until all master tasks have joined a global barrier: determining whether the task is a master task; if the task is not a master task, joining a single local barrier; if the task is a master task, joining the global barrier and the single local barrier only after all other tasks on the compute node have joined the single local barrier.
摘要翻译：在并行计算机上执行计算任务，其中包括耦合用于数据通信的计算节点，其中每个计算节点执行任务，每个计算节点上指定为主任务的一个任务包括：对于每个计算节点上的每个任务，直到所有主任务具有加入了全球壁垒：确定任务是否是主要任务; 如果任务不是主要任务，加入一个单一的局部障碍; 如果任务是主任务，则仅在计算节点上的所有其他任务已加入单个本地障碍之后才加入全局障碍和单个本地屏障。

7. 发明授权

US08909716B2 Administering truncated receive functions in a parallel messaging interface 有权
标题翻译：在并行消息接口中管理截断的接收功能
公开(公告)号：US08909716B2
公开(公告)日：2014-12-09
申请号：US12892153
申请日：2010-09-28
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F15/16
CPC分类号： G06F15/16
摘要： Administering truncated receive functions in a parallel messaging interface (‘PMI’) of a parallel computer comprising a plurality of compute nodes coupled for data communications through the PMI and through a data communications network, including: sending, through the PMI on a source compute node, a quantity of data from the source compute node to a destination compute node; specifying, by an application on the destination compute node, a portion of the quantity of data to be received by the application on the destination compute node and a portion of the quantity of data to be discarded; receiving, by the PMI on the destination compute node, all of the quantity of data; providing, by the PMI on the destination compute node to the application on the destination compute node, only the portion of the quantity of data to be received by the application; and discarding, by the PMI on the destination compute node, the portion of the quantity of data to be discarded.
摘要翻译：在并行计算机的并行消息接口（“PMI”）中管理截断的接收功能，所述并行计算机包括多个计算节点，所述计算节点被耦合用于通过所述PMI并通过数据通信网络进行数据通信，包括：通过源计算节点，从源计算节点到目的地计算节点的数据量; 由目的地计算节点上的应用指定要由目标计算节点上的应用接收的数据量的一部分和待丢弃的数据量的一部分; 由目标计算节点上的PMI接收所有数据量; 由目的地计算节点上的PMI向目标计算节点上的应用提供应用程序要接收的数据量的部分; 并且由目的地计算节点上的PMI舍弃待丢弃的数据量的部分。

8. 发明授权

US08656448B2 Providing policy-based application services to an application running on a computing system 失效
标题翻译：为运行在计算系统上的应用程序提供基于策略的应用程序服务
公开(公告)号：US08656448B2
公开(公告)日：2014-02-18
申请号：US11553101
申请日：2006-10-26
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Albert Sidelnik , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Albert Sidelnik , Brian E. Smith
IPC分类号： H04L29/06
CPC分类号： G06F11/3447 , G06F11/3409
摘要： Methods, apparatus, products are disclosed for providing policy-based application services to an application running on a computing system. The computing system includes at least one compute node. The compute node includes an application and a plurality of application services of a service type. Providing policy-based application services to an application running on a computing system includes establishing, on the compute node, a service policy specifying one of the application services of the service type for use by the application, establishing a performance ruleset, the performance ruleset specifying rules for analyzing the performance of the computing system, measuring performance of the computing system, and adjusting the service policy in dependence upon the performance ruleset and the measured performance. Providing policy-based application services to an application running on a computing system may also include accessing, by the application, the specified application service.
摘要翻译：公开了用于向在计算系统上运行的应用程序提供基于策略的应用服务的方法，装置，产品。计算系统包括至少一个计算节点。计算节点包括服务类型的应用和多个应用服务。向在计算系统上运行的应用提供基于策略的应用服务包括在计算节点上建立指定服务类型的应用服务之一以供应用使用的服务策略，建立性能规则集，指定性能规则集用于分析计算系统的性能，衡量计算系统的性能以及根据性能规则集和测量性能调整服务策略的规则。向运行在计算系统上的应用提供基于策略的应用服务也可以包括由应用访问指定的应用服务。

9. 发明申请

US20130304995A1 Scheduling Synchronization In Association With Collective Operations In A Parallel Computer 有权
标题翻译：在并行计算机中与集体操作相关联的调度同步
公开(公告)号：US20130304995A1
公开(公告)日：2013-11-14
申请号：US13470932
申请日：2012-05-14
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F12/00
CPC分类号： G06F15/17325
摘要： Methods, apparatuses, and computer program products for scheduling synchronization in association with collective operations in a parallel computer that includes a shared memory and a plurality of compute nodes that execute a parallel application utilizing the shared memory are provided. Embodiments include acquiring an available channel of the shared memory; posting to the acquired channel of the shared memory one or more collective operations and a synchronization point; determining that processing within the acquired channel has reached the synchronization point; and posting to the acquired channel, in response to determining that processing within the acquired channel has reached the synchronization point, a background synchronization operation corresponding to the one or more collective operations.
摘要翻译：提供了用于在并行计算机中与集合操作相关联的用于调度同步的方法，装置和计算机程序产品，所述并行计算机包括共享存储器和使用所述共享存储器执行并行应用的多个计算节点。实施例包括获取共享存储器的可用信道; 向共享存储器的获取通道发布一个或多个集合操作和同步点; 确定所获取的信道内的处理已经达到同步点; 并且响应于确定所获取的信道中的处理已经到达同步点，发布到获取的信道，对应于一个或多个集合操作的后台同步操作。

10. 发明授权

US08572629B2 Data communications in a parallel active messaging interface of a parallel computer 失效
标题翻译：并行计算机的并行活动消息接口中的数据通信
公开(公告)号：US08572629B2
公开(公告)日：2013-10-29
申请号：US12963694
申请日：2010-12-09
申请人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
发明人： Charles J. Archer , Michael A. Blocksome , Joseph D. Ratterman , Brian E. Smith
IPC分类号： G06F9/46
CPC分类号： G06F9/546
摘要： Data communications in a parallel active messaging interface (‘PAMI’) of a parallel computer, the parallel computer including a plurality of compute nodes that execute a parallel application, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a data communications instruction, the instruction characterized by an instruction type, the instruction specifying a transmission of transfer data from the origin endpoint to a target endpoint and transmitting, in accordance with the instruction type, the transfer data from the origin endpoint to the target endpoint.
摘要翻译：并行计算机的并行活动消息接口（“PAMI”）中的数据通信，并行计算机包括执行并行应用的多个计算节点，由数据通信端点组成的PAMI，每个端点包括数据通信参数的规范对于在计算节点上执行的线程，包括客户端，上下文和任务的规范，所述计算节点和端点被耦合用于通过PAMI进行数据通信，并且通过数据通信资源，包括在源终端中接收 PAMI数据通信指令，以指令类型为特征的指令，指示传输数据从原始端点传输到目标端点的指令，并根据指令类型将传输数据从原点终端发送到目标端点。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式