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. 发明申请

US20190197138A1 DATA SHUFFLING WITH HIERARCHICAL TUPLE SPACES 审中-公开
公开(公告)号：US20190197138A1
公开(公告)日：2019-06-27
申请号：US15851511
申请日：2017-12-21
申请人： International Business Machines Corporation
发明人： Abdullah Kayi , Carlos Henrique Andrade Costa , Yoonho Park , Charles Johns
IPC分类号： G06F17/30
CPC分类号： G06F16/212 , G06F16/2282 , G06F16/2379
摘要： Methods and systems for shuffling data to generate a dataset are described. A first map module may generate first pair data, and a second map module may generate second pair data, from source data. The first map module may insert the first pair data into a first local tuple space accessible to the first map module. The second map module may insert the second pair data into a second local tuple space accessible to the second map module. A shuffle module may request pair data that includes a particular key. The first and second pair data may be inserted into a global tuple space accessible by the first and second map modules. The shuffle module may identify the requested pair data in the global tuple space, and may fetch the identified pair data from a memory. The shuffle module may shuffle the fetched pair data to generate the dataset.

2. 发明申请

US20170358124A1 KERNEL CONVOLUTION FOR STENCIL COMPUTATION OPTIMIZATION 有权
公开(公告)号：US20170358124A1
公开(公告)日：2017-12-14
申请号：US15670756
申请日：2017-08-07
申请人： International Business Machines Corporation
发明人： Guilherme C. Januario , Yoonho Park , Bryan S. Rosenburg
IPC分类号： G06T15/10 , G06T15/00 , G06T5/50
CPC分类号： G06T15/10 , G06F17/16 , G06T5/50 , G06T15/005 , G06T2207/10004
摘要： Data processing techniques are provided to increase a computational speed of iterative computations that are performed over a domain of data points, such as stencil computations. For example, a method includes loading a set of domain data points into a cache memory; obtaining an iteration count T, and a base stencil operator having a first set of coefficients; generating a convolved stencil operator having a second set of coefficients, wherein the convolved stencil operator is generated by convolving the base stencil operator with itself at least one time; and iteratively processing the set of domain data points in the cache memory using the convolved stencil operator no more than T/2 iterations to obtain final processing results. The final processing results are similar to processing results that would be obtained by iteratively processing the set of domain data points using the base stencil operator for the iteration count T.

3. 发明授权

US09800691B2 Stream processing using a client-server architecture 有权
公开(公告)号：US09800691B2
公开(公告)日：2017-10-24
申请号：US14926983
申请日：2015-10-29
申请人： International Business Machines Corporation
发明人： Yoonho Park , Kenneth S. Sabir
IPC分类号： G06F15/16 , H04L29/06 , G06F17/30 , H04L29/08 , G06F21/44 , G06F21/73 , G06F9/50
CPC分类号： H04L67/42 , G06F9/5005 , G06F17/30516 , G06F21/44 , G06F21/73 , G06F2221/2103 , H04L65/605 , H04L65/607 , H04L67/02
摘要： Responsive to a client request, a processing thread for handling the client request is assigned. Responsive to the client request, a server request is sent to a stream server configured to interact with a plurality of stream processing nodes. The processing thread is maintained in an idle state pending a write response message from the stream server. The processing thread is returned to an active state responsive to receiving the write response message including a stream processing result from the stream server. A client response including the stream processing result is sent to the client.

4. 发明授权

US09317350B2 Method and apparatus for faulty memory utilization 有权
标题翻译：存储器利用不良的方法和装置
公开(公告)号：US09317350B2
公开(公告)日：2016-04-19
申请号：US14022171
申请日：2013-09-09
申请人： International Business Machines Corporation
发明人： Chen-Yong Cher , Carlos H. Andrade Costa , Yoonho Park , Bryan S. Rosenburg , Kyung D. Ryu
IPC分类号： G06F11/00 , G06F11/07
CPC分类号： G06F11/07 , G06F11/008 , G06F11/073 , G06F11/076 , G06F11/3037 , G06F11/3058 , G06F2201/81 , G06F2201/88
摘要： A method for faulty memory utilization in a memory system includes: obtaining information regarding memory health status of at least one memory page in the memory system; determining an error tolerance of the memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory system affecting the memory page; initiating a migration of data stored in the memory page when it is determined that the data stored in the memory page is non-error-tolerant; notifying at least one application regarding a predicted operating system failure and/or a predicted application failure when it is determined that data stored in the memory page is non-error-tolerant and cannot be migrated; and notifying at least one application regarding the memory failure predicted to occur when it is determined that data stored in the memory page is error-tolerant.
摘要翻译：一种存储器系统中存储器利用有缺陷的方法包括：获得关于存储器系统中的至少一个存储器页的存储器健康状态的信息; 当关于存储器健康状态的信息指示在存储器系统影响存储器页的区域中预测出现故障时，确定存储器页的容错; 当确定存储在存储器页面中的数据是不允许的时，发起存储在存储器页面中的数据的迁移; 当确定存储在存储器页面中的数据是不允许的且不能被迁移时，通知至少一个应用程序关于预测的操作系统故障和/或预测的应用程序故障; 并且当确定存储在存储器页面中的数据是容错的时，通知至少一个关于预测发生的存储器故障的应用程序。

5. 发明授权

US09298553B2 Methods, apparatus and system for selective duplication of subtasks 有权
标题翻译：用于选择性重复子任务的方法，装置和系统
公开(公告)号：US09298553B2
公开(公告)日：2016-03-29
申请号：US14176083
申请日：2014-02-08
申请人： International Business Machines Corporation
发明人： Carlos H. Andrade Costa , Chen-Yong Cher , Yoonho Park , Bryan S. Rosenburg , Kyung D. Ryu
IPC分类号： G06F11/00 , G06F11/14 , G06F11/07 , G06F11/20 , H04L29/08
CPC分类号： G06F11/1415 , G06F11/008 , G06F11/0754 , G06F11/1482 , G06F11/1492 , G06F11/2023 , G06F11/203 , G06F11/2048 , G06F11/302 , G06F11/3409 , H04L67/1034
摘要： A method for selective duplication of subtasks in a high-performance computing system includes: monitoring a health status of one or more nodes in a high-performance computing system, where one or more subtasks of a parallel task execute on the one or more nodes; identifying one or more nodes as having a likelihood of failure which exceeds a first prescribed threshold; selectively duplicating the one or more subtasks that execute on the one or more nodes having a likelihood of failure which exceeds the first prescribed threshold; and notifying a messaging library that one or more subtasks were duplicated.
摘要翻译：一种用于在高性能计算系统中选择性地复制子任务的方法包括：监视高性能计算系统中的一个或多个节点的健康状况，其中并行任务的一个或多个子任务在所述一个或多个节点上执行; 将一个或多个节点识别为具有超过第一规定阈值的故障可能性; 选择性地复制在具有超过第一规定阈值的故障可能性的一个或多个节点上执行的一个或多个子任务; 并通知消息传递库一个或多个子任务被复制。

6. 发明申请

US20210201130A1 JOB-LAUNCH TIME REDUCTION BY NODE PRE-CONFIGURATION 有权
公开(公告)号：US20210201130A1
公开(公告)日：2021-07-01
申请号：US16730401
申请日：2019-12-30
申请人： International Business Machines Corporation
发明人： Eun Kyung Lee , Giacomo Domeniconi , Alessandro Morari , Yoonho Park
IPC分类号： G06N3/08 , G06N3/04 , G06F9/445
摘要： Based on historic job data, a computer processor can predict a configuration of a computer node for running a future computer job. The computer processor can pre-configure the computer node based on the predicted configuration. Responsive to receiving a submission of a job, the computer processor can launch the job on the pre-configured computer node.

7. 发明授权

US10545839B2 Checkpointing using compute node health information 有权
公开(公告)号：US10545839B2
公开(公告)日：2020-01-28
申请号：US15853343
申请日：2017-12-22
申请人： International Business Machines Corporation
发明人： Carlos Henrique Andrade Costa , Yoonho Park , Chen-Yong Cher , Bryan Rosenburg , Kyung Ryu
IPC分类号： G06F11/00 , G06F11/14
摘要： A method is disclosed, as well as an associated apparatus and computer program product, for checkpointing using a plurality of communicatively coupled compute nodes. The method comprises acquiring health information for a first node of the plurality of compute nodes, and determining a first failure probability for the first node using the health information. The first failure probability corresponds to a predetermined time interval. The method further comprises selecting a second node of the plurality of compute nodes as a partner node for the first node. The second node has a second failure probability for the time interval. A composite failure probability of the first node and the second node is less than the first failure probability. The method further comprises copying checkpoint information from the first node to the partner node.

8. 发明申请

US20190163540A1 JOB SCHEDULING BASED ON NODE AND APPLICATION CHARACTERISTICS 审中-公开
公开(公告)号：US20190163540A1
公开(公告)日：2019-05-30
申请号：US15827208
申请日：2017-11-30
申请人： International Business Machines Corporation
发明人： Eun Kyung Lee , Bilge Acun , Yoonho Park
IPC分类号： G06F9/50 , G06F9/455
摘要： Aspects of the present invention disclose a method, computer program product, and system for scheduling an application. The method includes one or more processors receiving a task, the task includes instructions indicating desired nodes to perform the task through programs. The method further includes one or more processors identifying application characteristic information and node characteristic information associated with nodes within a data center composed of nodes. The application characteristic information includes resource utilization information for applications on nodes within the data center. The method further includes one or more processors determining that the nodes reach a threshold level of power consumption. The threshold level is a pre-set maximum amount of power utilized by a node within the data center. The method further includes one or more processors determining a node consuming an amount of power that is below a threshold level of power consumption in the data center.

9. 发明申请

US20190033944A1 POWER EFFICIENCY-AWARE NODE COMPONENT ASSEMBLY 审中-公开
公开(公告)号：US20190033944A1
公开(公告)日：2019-01-31
申请号：US15658494
申请日：2017-07-25
申请人： International Business Machines Corporation
发明人： Eun Kyung Lee , Bilge Acun , Yoonho Park
IPC分类号： G06F1/32
摘要： Sub-components assembled into a computer are selected based on sub-component power efficiency levels (for example, low, medium, high) and/or anticipated usage of the computer. Multiple units of each type of sub-component (for example, a CPU) are tested to determine a power efficiency level of each unit. Computers in which sub-component efficiency levels are desired to match an overall computer efficiency level, receive sub-component units of corresponding efficiency level. Computers anticipated to run applications that make intensive use of a given type of sub-component receive the given units having a higher efficiency level. Computers anticipated to run applications that make little use of a given type of sub-component receive a physical unit having a lower efficiency level. Computers anticipated to run a wide variety of applications of no particular usage intensity for a given type of sub-component, receive a unit having an average efficiency level.

10. 发明授权

US10073739B2 Methods, apparatus and system for selective duplication of subtasks 有权
公开(公告)号：US10073739B2
公开(公告)日：2018-09-11
申请号：US14957584
申请日：2015-12-02
申请人： International Business Machines Corporation
发明人： Carlos H. Andrade Costa , Chen-Yong Cher , Yoonho Park , Bryan S. Rosenburg , Kyung D. Ryu
IPC分类号： G06F11/00 , G06F11/14 , G06F11/07 , G06F11/20 , H04L29/08 , G06F11/30 , G06F11/34
CPC分类号： G06F11/1415 , G06F11/008 , G06F11/0754 , G06F11/1482 , G06F11/1492 , G06F11/2023 , G06F11/203 , G06F11/2048 , G06F11/302 , G06F11/3409 , H04L67/1034
摘要： A method for selective duplication of subtasks in a high-performance computing system includes: monitoring a health status of one or more nodes in a high-performance computing system, where one or more subtasks of a parallel task execute on the one or more nodes; identifying one or more nodes as having a likelihood of failure which exceeds a first prescribed threshold; selectively duplicating the one or more subtasks that execute on the one or more nodes having a likelihood of failure which exceeds the first prescribed threshold; and notifying a messaging library that one or more subtasks were duplicated.

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