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

US20120239909A1 SYSTEMS AND METHODS FOR VOTING AMONG PARALLEL THREADS 审中-公开
标题翻译：用于表示并行线程的系统和方法
公开(公告)号：US20120239909A1
公开(公告)日：2012-09-20
申请号：US13485622
申请日：2012-05-31
申请人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
发明人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
IPC分类号： G06F9/00
CPC分类号： G06F9/3851 , G06F9/30087 , G06F9/3009 , G06F9/3887
摘要： One embodiment of the present invention sets forth a technique for efficiently performing voting operations within a multi-threaded parallel-processing system. A group of related parallel program threads executes within a processor core together in parallel. A new instruction, called a “vote” instruction, is introduced that enables a parallel program thread to post an individual vote within the context of the group of related threads and to receive the result of the vote. In this fashion, the vote instruction advantageously reduces overhead associated with inter-thread communication, thereby improving overall system performance.
摘要翻译：本发明的一个实施例提出了一种用于在多线程并行处理系统内有效执行投票操作的技术。一组相关的并行程序线程并行执行在处理器内核中。引入了一项称为“投票”指令的新指令，使得并行程序线程能够在相关线程组的上下文中发布个人投票并接收投票结果。以这种方式，投票指令有利地减少与线程间通信相关联的开销，从而提高整体系统性能。

2. 发明授权

US08200947B1 Systems and methods for voting among parallel threads 有权
标题翻译：并行线程中投票的系统和方法
公开(公告)号：US08200947B1
公开(公告)日：2012-06-12
申请号：US12054322
申请日：2008-03-24
申请人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
发明人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
IPC分类号： G06F9/00
CPC分类号： G06F9/3851 , G06F9/30087 , G06F9/3009 , G06F9/3887
摘要： One embodiment of the present invention sets forth a technique for efficiently performing voting operations within a multi-threaded parallel-processing system. A group of related parallel program threads executes within a processor core together in parallel. A new instruction, called a “vote” instruction, is introduced that enables a parallel program thread to post an individual vote within the context of the group of related threads and to receive the result of the vote. In this fashion, the vote instruction advantageously reduces overhead associated with inter-thread communication, thereby improving overall system performance.
摘要翻译：本发明的一个实施例提出了一种用于在多线程并行处理系统内有效执行投票操作的技术。一组相关的并行程序线程并行执行在处理器内核中。引入了一项称为“投票”指令的新指令，使得并行程序线程能够在相关线程组的上下文中发布个人投票并接收投票结果。以这种方式，投票指令有利地减少与线程间通信相关联的开销，从而提高整体系统性能。

3. 发明授权

US08214625B1 Systems and methods for voting among parallel threads 有权
公开(公告)号：US08214625B1
公开(公告)日：2012-07-03
申请号：US12324645
申请日：2008-11-26
申请人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
发明人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
IPC分类号： G06F9/00
CPC分类号： G06F9/3851 , G06F9/30087 , G06F9/3009 , G06F9/3887
摘要： One embodiment of the present invention sets forth a technique for efficiently performing voting operations within a multi-threaded parallel-processing system. A group of related parallel program threads executes within a processor core together in parallel. A new instruction, called a “vote” instruction, is introduced that enables a parallel program thread to post an individual vote within the context of the group of related threads and to receive the result of the vote. In this fashion, the vote instruction advantageously reduces overhead associated with inter-thread communication, thereby improving overall system performance.

4. 发明授权

US10152328B2 Systems and methods for voting among parallel threads 有权
公开(公告)号：US10152328B2
公开(公告)日：2018-12-11
申请号：US13485622
申请日：2012-05-31
申请人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
发明人： John R. Nickolls , Lars Nyland , Peter C. Mills , Jeremy Sugerman , Timothy Foley , Brian Fahs , Michael Garland , David P. Luebke
IPC分类号： G06F9/38 , G06F9/30
摘要： One embodiment of the present invention sets forth a technique for efficiently performing voting operations within a multi-threaded parallel-processing system. A group of related parallel program threads executes within a processor core together in parallel. A new instruction, called a “vote” instruction, is introduced that enables a parallel program thread to post an individual vote within the context of the group of related threads and to receive the result of the vote. In this fashion, the vote instruction advantageously reduces overhead associated with inter-thread communication, thereby improving overall system performance.

5. 发明授权

US08281318B2 Systems and methods for inter process communication based on queues 有权
标题翻译：基于队列的进程间通信的系统和方法
公开(公告)号：US08281318B2
公开(公告)日：2012-10-02
申请号：US12167141
申请日：2008-07-02
申请人： Anupam Chanda , Kevin Scott Christopher , Jeremy Sugerman , Petr Vandrovec , Gustav Seth Wibling
发明人： Anupam Chanda , Kevin Scott Christopher , Jeremy Sugerman , Petr Vandrovec , Gustav Seth Wibling
IPC分类号： G06F3/00
CPC分类号： G06F9/546 , G06F9/45533 , G06F9/4812 , G06F2209/481 , G06F2209/548
摘要： A method of data communication between a first virtual machine and a second virtual machine is disclosed. The second virtual machine is executing in a record/replay mode. Data from the first virtual machine is copied to a first queue. The first queue receives the data from the first virtual machine. The first queue has a header section and a data section, wherein the header section being write protected and stores a tail pointer of the data in the first queue. The tail pointer is updated in the header section. This update of the tail pointer causes a page fault, which is handled-through a page fault handler. The handling includes copying the data from the first queue to a second queue. The second queue being configured to receive a copy of the data and to allow the second virtual machine to access the copy of the data.
摘要翻译：公开了第一虚拟机与第二虚拟机之间的数据通信的方法。第二个虚拟机正在记录/重播模式下执行。来自第一个虚拟机的数据被复制到第一个队列。第一个队列从第一个虚拟机接收数据。第一队列具有报头部分和数据部分，其中报头部分被写保护，并将数据的尾部指针存储在第一队列中。尾部指针在标题部分更新。尾部指针的这种更新会导致页面错误，这是通过页面错误处理程序处理的。处理包括将数据从第一个队列复制到第二个队列。第二队列被配置为接收数据的副本并且允许第二虚拟机访问数据的副本。

6. 发明授权

US08732700B2 Virtualization system with a remote proxy 有权
标题翻译：具有远程代理的虚拟化系统
公开(公告)号：US08732700B2
公开(公告)日：2014-05-20
申请号：US12337732
申请日：2008-12-18
申请人： Matthew David Ginzton , Jeremy Sugerman , Ramesh Dharan
发明人： Matthew David Ginzton , Jeremy Sugerman , Ramesh Dharan
IPC分类号： G06F9/455 , G06F9/46 , G06F15/173
CPC分类号： G06F9/455 , G06F9/45533 , G06F9/45558 , G06F9/50 , G06F9/5005 , G06F9/5061 , G06F9/5077 , G06F2009/45595
摘要： A remote proxy provides remote control to a virtualization system. The virtualization system includes a computer having virtualization software executing on the computer, and one or more virtual machines executing on the computer by way of the virtualization software, each of the virtual machines having a guest operating system. The remote proxy receives over a network connection a plurality of logical channels of a multi-channel remote control protocol at a first domain of a plurality of domains of the virtualization system. Each of the logical channels convey information corresponding to one of the domains. For each logical channel, the remote proxy identifies the corresponding domain. Further, for each of the logical channels corresponding to a domain other than the first domain, the remote proxy forwards the logical channel to the corresponding domain.
摘要翻译：远程代理为虚拟化系统提供远程控制。虚拟化系统包括具有在计算机上执行的虚拟化软件的计算机，以及通过虚拟化软件在计算机上执行的一个或多个虚拟机，每个虚拟机具有来宾操作系统。远程代理在虚拟化系统的多个域的第一域处通过网络连接接收多信道远程控制协议的多个逻辑信道。每个逻辑信道传送与其中一个域对应的信息。对于每个逻辑信道，远程代理识别相应的域。此外，对于与第一域以外的域相对应的每个逻辑信道，远程代理将逻辑信道转发到对应的域。

7. 发明授权

US09104449B2 Optimized execution of dynamic languages 有权
标题翻译：优化动态语言的执行
公开(公告)号：US09104449B2
公开(公告)日：2015-08-11
申请号：US13526328
申请日：2012-06-18
申请人： Solomon Boulos , Jeremy Sugerman
发明人： Solomon Boulos , Jeremy Sugerman
IPC分类号： G06F9/45 , G06F9/455
CPC分类号： G06F9/45508
摘要： Program code in a dynamic language is executed through a dynamic execution module that includes an execution context and an execution codebase. The instructions of the program code are translated into machine code that is stored in the execution codebase, and the execution context tracks the runtime state and persistent state of the program through executions and across executions. Using the execution codebase and the execution context the dynamic execution module minimizes duplicative machine code generation, while preserving the dynamic aspects of the program.
摘要翻译：动态语言中的程序代码通过包含执行上下文和执行代码库的动态执行模块执行。程序代码的指令被转换为存储在执行代码库中的机器代码，并且执行上下文通过执行和跨执行跟踪程序的运行时状态和持续状态。使用执行代码库和执行上下文，动态执行模块将重复的机器代码生成最小化，同时保留程序的动态方面。

8. 发明申请

US20130339936A1 Optimized Execution of Dynamic Languages 有权
标题翻译：动态语言的优化执行
公开(公告)号：US20130339936A1
公开(公告)日：2013-12-19
申请号：US13526328
申请日：2012-06-18
申请人： Solomon Boulos , Jeremy Sugerman
发明人： Solomon Boulos , Jeremy Sugerman
IPC分类号： G06F9/45
CPC分类号： G06F9/45508
摘要： Program code in a dynamic language is executed through a dynamic execution module that includes an execution context and an execution codebase. The instructions of the program code are translated into machine code that is stored in the execution codebase, and the execution context tracks the runtime state and persistent state of the program through executions and across executions. Using the execution codebase and the execution context the dynamic execution module minimizes duplicative machine code generation, while preserving the dynamic aspects of the program.
摘要翻译：动态语言中的程序代码通过包含执行上下文和执行代码库的动态执行模块执行。程序代码的指令被转换为存储在执行代码库中的机器代码，并且执行上下文通过执行和跨执行跟踪程序的运行时状态和持续状态。使用执行代码库和执行上下文，动态执行模块将重复的机器代码生成最小化，同时保留程序的动态方面。

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