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

US20130191612A1 INTERFERENCE-DRIVEN RESOURCE MANAGEMENT FOR GPU-BASED HETEROGENEOUS CLUSTERS 有权
标题翻译：基于GPU的异构集群的干扰驱动资源管理
公开(公告)号：US20130191612A1
公开(公告)日：2013-07-25
申请号：US13646661
申请日：2012-10-06
申请人： Cheng-Hong Li , Srihari Cadambi , Srimat T. Chakradhar , Rajat Phull
发明人： Cheng-Hong Li , Srihari Cadambi , Srimat T. Chakradhar , Rajat Phull
IPC分类号： G06F15/76
CPC分类号： G06T15/005 , G06F2209/501 , G06T2210/52
摘要： Systems and methods are disclosed that share coprocessor resources between two or more applications in a computing cluster using a job selector to receive jobs from a job queue; a node selector coupled to the job selector; an off line profiler with an interference prediction model; a coprocessor dynamic interference detection module; and a coprocessor interference response module.
摘要翻译：公开了系统和方法，其使用作业选择器在计算集群中的两个或更多应用之间共享协处理器资源以从作业队列接收作业; 耦合到作业选择器的节点选择器; 具有干扰预测模型的离线轮廓仪; 协处理器动态干扰检测模块; 和协处理器干扰响应模块。

2. 发明授权

US09135741B2 Interference-driven resource management for GPU-based heterogeneous clusters 有权
标题翻译：基于GPU的异构集群的干扰驱动资源管理
公开(公告)号：US09135741B2
公开(公告)日：2015-09-15
申请号：US13646661
申请日：2012-10-06
申请人： Cheng-Hong Li , Srihari Cadambi , Srimat T Chakradhar , Rajat Phull
发明人： Cheng-Hong Li , Srihari Cadambi , Srimat T Chakradhar , Rajat Phull
IPC分类号： G06F15/00 , G06F15/76 , G06T15/00
CPC分类号： G06T15/005 , G06F2209/501 , G06T2210/52
摘要： Systems and methods are disclosed that share coprocessor resources between two or more applications in a computing cluster using a job selector to receive jobs from a job queue; a node selector coupled to the job selector; an off line profiler with an interference prediction model; a coprocessor dynamic interference detection module; and a coprocessor interference response module.
摘要翻译：公开了系统和方法，其使用作业选择器在计算集群中的两个或更多应用之间共享协处理器资源以从作业队列接收作业; 耦合到作业选择器的节点选择器; 具有干扰预测模型的离线轮廓仪; 协处理器动态干扰检测模块; 和协处理器干扰响应模块。

3. 发明申请

US20120290283A1 Workload Partitioning Procedure For Null Message-Based PDES 有权
标题翻译：基于消息的消息PDES的工作负载分区过程
公开(公告)号：US20120290283A1
公开(公告)日：2012-11-15
申请号：US13107110
申请日：2011-05-13
申请人： Cheng-Hong Li , Alfred J. Park , Eugen Schenfeld
发明人： Cheng-Hong Li , Alfred J. Park , Eugen Schenfeld
IPC分类号： G06F9/45
CPC分类号： G06F9/5077 , G06F9/45558 , G06F17/5022 , G06F2009/45595
摘要： An embodiment of the invention includes applying a first partition to a plurality of LPs, wherein a particular LP is assigned to a first set of LPs. A second partition is applied to the LPs, wherein the particular LP is assigned to an LP set different from the first set. For both the first and second partitions, lookahead values and transit times are determined for each of the LPs and related links. For the first partition, a first system progression rate is computed using a specified function with the lookahead values and transit times determined for the first partition. For the second partition, a second system progression rate is computed using the specified function with the lookahead values and transit times determined for the second partition. The first and second system progression rates are compared to determine which is the lowest.
摘要翻译：本发明的实施例包括将第一分区应用于多个LP，其中特定LP被分配给第一组LP。将第二分区应用于LP，其中特定LP被分配给不同于第一组的LP集合。对于第一和第二分区，为每个LP和相关链路确定查找值和传输时间。对于第一分区，使用具有为第一分区确定的前瞻值和传输时间的指定函数计算第一系统进度。对于第二分区，使用指定的函数计算第二系统进展速率，其中为第二分区确定了前瞻值和传送时间。比较第一和第二系统进展率以确定哪个是最低的。

4. 发明授权

US09798587B2 Workload partitioning procedure for null message-based PDES 有权
公开(公告)号：US09798587B2
公开(公告)日：2017-10-24
申请号：US13107110
申请日：2011-05-13
申请人： Cheng-Hong Li , Alfred J. Park , Eugen Schenfeld
发明人： Cheng-Hong Li , Alfred J. Park , Eugen Schenfeld
IPC分类号： G06F9/50 , G06F17/50 , G06F9/455
CPC分类号： G06F9/5077 , G06F9/45558 , G06F17/5022 , G06F2009/45595
摘要： An embodiment of the invention includes applying a first partition to a plurality of LPs, wherein a particular LP is assigned to a first set of LPs. A second partition is applied to the LPs, wherein the particular LP is assigned to an LP set different from the first set. For both the first and second partitions, lookahead values and transit times are determined for each of the LPs and related links. For the first partition, a first system progression rate is computed using a specified function with the lookahead values and transit times determined for the first partition. For the second partition, a second system progression rate is computed using the specified function with the lookahead values and transit times determined for the second partition. The first and second system progression rates are compared to determine which is the lowest.

5. 发明授权

US08875108B2 Run-time bottleneck detection 有权
标题翻译：运行时瓶颈检测
公开(公告)号：US08875108B2
公开(公告)日：2014-10-28
申请号：US13349357
申请日：2012-01-12
申请人： Yoonseo Choi , Cheng-Hong Li
发明人： Yoonseo Choi , Cheng-Hong Li
IPC分类号： G06F9/44 , G06F15/16 , G06F15/173
CPC分类号： G06F11/3612 , G06F11/3419 , G06F2201/865
摘要： A method of detecting a bottleneck in a data flow program includes executing the data flow program, wherein the data flow program comprises a kernel, determining a percentage of time the kernel spends on a computation, and determining the kernel to be a bottleneck upon comparing the percentage of time the kernel spends on the computation to a condition.
摘要翻译：检测数据流程序中的瓶颈的方法包括执行数据流程序，其中数据流程序包括内核，确定内核花费在计算上的时间百分比，以及在比较内核时将内核确定为瓶颈内核花费在计算到条件的时间百分比。

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