专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

41. 发明申请

US20090055474A1 Line-Plane Broadcasting in a Data Communications Network of a Parallel Computer 失效
公开(公告)号：US20090055474A1
公开(公告)日：2009-02-26
申请号：US11843090
申请日：2007-08-22
申请人： Charles J. Archer , Jeremy E. Berg , Michael A. Blocksome , Brian E. Smith
发明人： Charles J. Archer , Jeremy E. Berg , Michael A. Blocksome , Brian E. Smith
IPC分类号： G06F15/173
CPC分类号： G06F15/173
摘要： Methods, apparatus, and products are disclosed for line-plane broadcasting in a data communications network of a parallel computer, the parallel computer comprising a plurality of compute nodes connected together through the network, the network optimized for point to point data communications and characterized by at least a first dimension, a second dimension, and a third dimension, that include: initiating, by a broadcasting compute node, a broadcast operation, including sending a message to all of the compute nodes along an axis of the first dimension for the network; sending, by each compute node along the axis of the first dimension, the message to all of the compute nodes along an axis of the second dimension for the network; and sending, by each compute node along the axis of the second dimension, the message to all of the compute nodes along an axis of the third dimension for the network.

42. 发明申请

US20080313376A1 Heuristic Status Polling 有权
标题翻译：启发式状态轮询
公开(公告)号：US20080313376A1
公开(公告)日：2008-12-18
申请号：US11764282
申请日：2007-06-18
申请人： Charles J. Archer , Michael A. Blocksome , Philip Heidelberger , Sameer Kumar , Jeffrey J. Parker , Joseph D. Ratterman
发明人： Charles J. Archer , Michael A. Blocksome , Philip Heidelberger , Sameer Kumar , Jeffrey J. Parker , Joseph D. Ratterman
IPC分类号： G06F13/22
CPC分类号： H04L12/44 , H04L12/403
摘要： Methods, compute nodes, and computer program products are provided for heuristic status polling of a component in a computing system. Embodiments include receiving, by a polling module from a requesting application, a status request requesting status of a component; determining, by the polling module, whether an activity history for the component satisfies heuristic polling criteria; polling, by the polling module, the component for status if the activity history for the component satisfies the heuristic polling criteria; and not polling, by the polling module, the component for status if the activity history for the component does not satisfy the heuristic criteria.
摘要翻译：提供方法，计算节点和计算机程序产品用于计算系统中组件的启发状态轮询。实施例包括通过轮询模块从请求应用程序接收请求状态的组件的状态请求; 由轮询模块确定该组件的活动历史是否满足启发式轮询标准; 如果组件的活动历史满足启发式轮询标准，轮询由轮询模块组成状态; 如果组件的活动历史不满足启发式标准，则轮询模块不会轮询该组件的状态。

43. 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。

44. 发明申请

US20080281997A1 Low Latency, High Bandwidth Data Communications Between Compute Nodes in a Parallel Computer 失效
标题翻译：并行计算机中计算节点之间的低延迟，高带宽数据通信
公开(公告)号：US20080281997A1
公开(公告)日：2008-11-13
申请号：US11746333
申请日：2007-05-09
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： G06F13/28
CPC分类号： G06F13/4269
摘要： Methods, parallel computers, and computer program products are disclosed for low latency, high bandwidth data communications between compute nodes in a parallel computer. Embodiments include receiving, by an origin direct memory access (‘DMA’) engine of an origin compute node, data for transfer to a target compute node; sending, by the origin DMA engine of the origin compute node to a target DMA engine on the target compute node, a request to send (‘RTS’) message; transferring, by the origin DMA engine, a predetermined portion of the data to the target compute node using memory FIFO operation; determining, by the origin DMA engine whether an acknowledgement of the RTS message has been received from the target DMA engine; if the an acknowledgement of the RTS message has not been received, transferring, by the origin DMA engine, another predetermined portion of the data to the target compute node using a memory FIFO operation; and if the acknowledgement of the RTS message has been received by the origin DMA engine, transferring, by the origin DMA engine, any remaining portion of the data to the target compute node using a direct put operation.
摘要翻译：公开了并行计算机和计算机程序产品的方法，用于并行计算机中的计算节点之间的低延迟，高带宽数据通信。实施例包括通过原始计算节点的原始直接存储器访问（“DMA”）引擎接收用于传送到目标计算节点的数据; 由原始计算节点的原始DMA引擎发送到目标计算节点上的目标DMA引擎，发送（'RTS'）消息的请求; 由原始DMA引擎使用存储器FIFO操作将预定部分的数据传送到目标计算节点; 由原始DMA引擎确定是否从目标DMA引擎接收到RTS消息的确认; 如果尚未接收到RTS消息的确认，则由原始DMA引擎使用存储器FIFO操作将另一预定部分的数据传送到目标计算节点; 并且如果原始DMA引擎已经接收到RTS消息的确认，则由原始DMA引擎使用直接放置操作将数据的剩余部分传送到目标计算节点。

45. 发明申请

US20080259916A1 OPPORTUNISTIC QUEUEING INJECTION STRATEGY FOR NETWORK LOAD BALANCING 有权
标题翻译：网络负载平衡机动队列注入策略
公开(公告)号：US20080259916A1
公开(公告)日：2008-10-23
申请号：US11738034
申请日：2007-04-20
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： H04L12/28
CPC分类号： H04L45/00 , H04L45/24 , H04L47/10 , H04L47/125
摘要： Embodiments of the invention include a method, system, and article of manufacture that provide opportunistic queuing injection strategy used for data communication between nodes of a parallel computer system. A message may be encapsulated into a set of data packets. When the packets are sent, an opportunistic injection queue may be configured to transmit them to multiple hardware injection ports. This approach allows for complete network link saturation. In a parallel system with network links in multiple dimensions, sending message packets using more than one dimension may substantially increase network throughput.
摘要翻译：本发明的实施例包括提供用于并行计算机系统的节点之间的数据通信的机会排队注入策略的方法，系统和制品。消息可以被封装到一组数据分组中。当发送数据包时，可以配置机会性注入队列将其发送到多个硬件注入端口。这种方法允许完整的网络链路饱和。在具有多个维度的网络链路的并行系统中，使用多个维度发送消息分组可以显着增加网络吞吐量。

46. 发明申请

US20080195840A1 Identifying Messaging Completion on a Parallel Computer 失效
标题翻译：识别并行计算机上的消息完成
公开(公告)号：US20080195840A1
公开(公告)日：2008-08-14
申请号：US11672989
申请日：2007-02-09
申请人： Charles J. Archer , Camesha R. Hardwick , Patrick J. McCarthy , Brian P. Wallenfelt
发明人： Charles J. Archer , Camesha R. Hardwick , Patrick J. McCarthy , Brian P. Wallenfelt
IPC分类号： G06F15/76 , G06F9/06
CPC分类号： G06F15/17337
摘要： Methods, parallel computers, and products are provided for identifying messaging completion on a parallel computer. The parallel computer includes a plurality of compute nodes, the compute nodes coupled for data communications by at least two independent data communications networks including a binary tree data communications network optimal for collective operations that organizes the nodes as a tree and a torus data communications network optimal for point to point operations that organizes the nodes as a torus. Embodiments include reading all counters at each node of the torus data communications network; calculating at each node a current node value in dependence upon the values read from the counters at each node; and determining for all nodes whether the current node value for each node is the same as a previously calculated node value for each node. If the current node is the same as the previously calculated node value for all nodes of the torus data communications network, embodiments include determining that messaging is complete and if the current node is not the same as the previously calculated node value for all nodes of the torus data communications network, embodiments include determining that messaging is currently incomplete.
摘要翻译：提供方法，并行计算机和产品用于标识并行计算机上的消息完成。并行计算机包括多个计算节点，所述计算节点被耦合用于由至少两个独立的数据通信网络进行数据通信，所述至少两个独立数据通信网络包括最佳的用于将节点组织为树的二进制树数据通信网络和圆环数据通信网络最优用于将节点组织为环面的点对点操作。实施例包括读取环面数据通信网络的每个节点处的所有计数器; 根据从每个节点处的计数器读取的值，在每个节点计算当前节点值; 以及为所有节点确定每个节点的当前节点值是否与每个节点的先前计算的节点值相同。如果当前节点与圆环数据通信网络的所有节点的先前计算的节点值相同，则实施例包括确定消息传递完成，并且如果当前节点与先前计算出的节点的所有节点的节点值不相同环面数据通信网络，实施例包括确定消息传递当前不完整。

47. 发明申请

US20080126739A1 Parallel Execution of Operations for a Partitioned Binary Radix Tree on a Parallel Computer 有权
标题翻译：并行计算机上并行执行分区二进制基树的操作
公开(公告)号：US20080126739A1
公开(公告)日：2008-05-29
申请号：US11531846
申请日：2006-09-14
申请人： Charles J. Archer , Benjamin E. Lynam , Gary R. Ricard
发明人： Charles J. Archer , Benjamin E. Lynam , Gary R. Ricard
IPC分类号： G06F12/00
CPC分类号： G06F17/30327 , G06F17/30445 , Y10S707/99937
摘要： Methods, apparatus, and products are disclosed for parallel execution of operations for a partitioned binary radix tree that include: receiving, in a parallel computer, an operational entry for the PBRT, the PBRT comprising a plurality of logical pages that contain a plurality of entries, each logical page included in a tier and containing one or more subentries corresponding to the tier of the logical page containing the subentry, each entry is composed of a subentry from each logical page on an entry path; processing in parallel, on the parallel computer, each logical page in each tier, including: identifying a portion of the operational entry that corresponds to the tier of the logical page, and performing an operation on the logical page in dependence upon the identified portion of the operational entry for the tier; and selecting operation results from the logical pages on the entry path for the operational entry.
摘要翻译：公开了用于并行执行分区二进制基树的操作的方法，装置和产品，包括：在并行计算机中接收PBRT的操作条目，PBRT包括包含多个条目的多个逻辑页面包含在层中并且包含与包含子条目的逻辑页的层相对应的一个或多个子条目的每个逻辑页面，每个条目由入口路径上每个逻辑页面的子条目组成; 在并行计算机上并行处理每层中的每个逻辑页面，包括：识别对应于逻辑页面层的操作条目的一部分，以及根据所识别的部分的逻辑页面对逻辑页面执行操作层次的操作入口; 以及从用于操作条目的入口路径上的逻辑页面中选择操作结果。

48. 发明申请

US20080072101A1 Identifying Failure in a Tree Network of a Parallel Computer 有权
标题翻译：识别并行计算机的树网络中的故障
公开(公告)号：US20080072101A1
公开(公告)日：2008-03-20
申请号：US11531787
申请日：2006-09-14
申请人： Charles J. Archer , Kurt W. Pinnow , Brian P. Wallenfelt
发明人： Charles J. Archer , Kurt W. Pinnow , Brian P. Wallenfelt
IPC分类号： G06F11/00
CPC分类号： G06F11/3409 , G06F11/2236 , G06F11/3485 , G06F2201/81
摘要： Methods, parallel computers, and products are provided for identifying failure in a tree network of a parallel computer. The parallel computer includes one or more processing sets including an I/O node and a plurality of compute nodes. For each processing set embodiments include selecting a set of test compute nodes, the test compute nodes being a subset of the compute nodes of the processing set; measuring the performance of the I/O node of the processing set; measuring the performance of the selected set of test compute nodes; calculating a current test value in dependence upon the measured performance of the I/O node of the processing set, the measured performance of the set of test compute nodes, and a predetermined value for I/O node performance; and comparing the current test value with a predetermined tree performance threshold. If the current test value is below the predetermined tree performance threshold, embodiments include selecting another set of test compute nodes. If the current test value is not below the predetermined tree performance threshold, embodiments include selecting from the test compute nodes one or more potential problem nodes and testing individually potential problem nodes and links to potential problem nodes.
摘要翻译：提供方法，并行计算机和产品用于识别并行计算机的树形网络中的故障。并行计算机包括一个或多个包括I / O节点和多个计算节点的处理集合。对于每个处理集合，实施例包括选择一组测试计算节点，测试计算节点是处理集合的计算节点的子集; 测量处理集的I / O节点的性能; 测量所选择的一组测试计算节点的性能; 根据测量的处理集合的I / O节点的性能，测试计算节点集合的测量性能以及I / O节点性能的预定值来计算当前测试值; 以及将当前测试值与预定树性能阈值进行比较。如果当前测试值低于预定树性能阈值，则实施例包括选择另一组测试计算节点。如果当前测试值不低于预定树性能阈值，则实施例包括从测试计算节点选择一个或多个潜在问题节点，并单独测试潜在问题节点和到潜在问题节点的链路。

49. 发明授权

US09495135B2 Developing collective operations for a parallel computer 有权
公开(公告)号：US09495135B2
公开(公告)日：2016-11-15
申请号：US13369451
申请日：2012-02-09
申请人： Charles J. Archer , James E. Carey , Philip J. Sanders , Brian E. Smith
发明人： Charles J. Archer , James E. Carey , Philip J. Sanders , Brian E. Smith
IPC分类号： G06F9/44 , G06F9/54
CPC分类号： G06F8/34 , G06F9/542
摘要： Developing collective operations for a parallel computer that includes compute nodes includes: presenting, by a collective development tool, a graphical user interface (‘GUI’) to a collective developer; receiving, by the collective development tool from the collective developer through the GUI, a selection of one or more collective primitives; receiving, by the collective development tool from the collective developer through the GUI, a specification of a serial order of the collective primitives and a specification of input and output buffers for each collective primitive; and generating, by the collective development tool in dependence upon the selection of collective primitives, the serial order of the collective primitives, and the input and output buffers for each collective primitive, executable code that carries out the collective operation specified by the collective primitives.

50. 发明授权

US09417905B2 Terminating an accelerator application program in a hybrid computing environment 有权
标题翻译：在混合计算环境中终止加速器应用程序
公开(公告)号：US09417905B2
公开(公告)日：2016-08-16
申请号：US12699162
申请日：2010-02-03
申请人： Charles J. Archer , Gregory H. Bellows , Dean J. Burdick , James E. Carey , Jeffrey M. Ceason , Matthew W. Markland , Philip J. Sanders , Gordon G. Stewart
发明人： Charles J. Archer , Gregory H. Bellows , Dean J. Burdick , James E. Carey , Jeffrey M. Ceason , Matthew W. Markland , Philip J. Sanders , Gordon G. Stewart
IPC分类号： G06F3/00 , G06F9/44 , G06F9/46 , G06F13/00
CPC分类号： G06F9/46
摘要： Terminating an accelerator application program in a hybrid computing environment that includes a host computer having a host computer architecture and an accelerator having an accelerator architecture, where the host computer and the accelerator are adapted to one another for data communications by a system level message passing module (‘SLMPM’), and terminating an accelerator application program in a hybrid computing environment includes receiving, by the SLMPM from a host application executing on the host computer, a request to terminate an accelerator application program executing on the accelerator; terminating, by the SLMPM, execution of the accelerator application program; returning, by the SLMPM to the host application, a signal indicating that execution of the accelerator application program was terminated; and performing, by the SLMPM, a cleanup of the execution environment associated with the terminated accelerator application program.
摘要翻译：在包括具有主机结构的主计算机和具有加速器架构的加速器的混合计算环境中终止加速器应用程序，其中所述主计算机和所述加速器彼此适配以用于由系统级消息传递模块进行数据通信（“SLMPM”），并且在混合计算环境中终止加速器应用程序包括：通过SLMPM从在主计算机上执行的主机应用程序接收终止在加速器上执行的加速器应用程序的请求; 通过SLMPM终止加速器应用程序的执行; 通过SLMPM向主机应用返回指示加速器应用程序的执行被终止的信号; 并且通过SLMPM执行与终止的加速器应用程序相关联的执行环境的清理。

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