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

US08392665B2 Allocation and write policy for a glueless area-efficient directory cache for hotly contested cache lines 有权
标题翻译：一个无缝区域高效的目录缓存的分配和写入策略，用于热烈争议的缓存行
公开(公告)号：US08392665B2
公开(公告)日：2013-03-05
申请号：US12890649
申请日：2010-09-25
申请人： Adrian C. Moga , Malcolm Mandviwalla , Vedaraman Geetha , Herbert H. Hum
发明人： Adrian C. Moga , Malcolm Mandviwalla , Vedaraman Geetha , Herbert H. Hum
IPC分类号： G06F12/00
CPC分类号： G06F12/0831 , G06F12/082 , G06F2212/2542
摘要： Methods and apparatus relating to allocation and/or write policy for a glueless area-efficient directory cache for hotly contested cache lines are described. In one embodiment, a directory cache stores data corresponding to a caching status of a cache line. The caching status of the cache line is stored for each of a plurality of caching agents in the system. An write-on-allocate policy is used for the directory cache by using a special state (e.g., snoop-all state) that indicates one or more snoops are to be broadcasted to all agents in the system. Other embodiments are also disclosed.
摘要翻译：描述与用于热挑战的高速缓存行的无胶带区域高效目录高速缓存的分配和/或写入策略有关的方法和装置。在一个实施例中，目录高速缓存存储与高速缓存行的高速缓存状态对应的数据。为系统中的多个缓存代理中的每一个存储缓存行的高速缓存状态。通过使用指示要广播到系统中的所有代理的一个或多个窥探的特殊状态（例如，窥探全状态），对目录高速缓存使用写入分配策略。还公开了其他实施例。

2. 发明授权

US08041898B2 Method, system and apparatus for reducing memory traffic in a distributed memory system 有权
标题翻译：用于减少分布式存储器系统中的存储器流量的方法，系统和装置
公开(公告)号：US08041898B2
公开(公告)日：2011-10-18
申请号：US12113268
申请日：2008-05-01
申请人： Adrian Moga , Rajat Agarwal , Malcolm Mandviwalla
发明人： Adrian Moga , Rajat Agarwal , Malcolm Mandviwalla
IPC分类号： G06F12/08
CPC分类号： G06F12/082 , G06F2212/1041
摘要： The present disclosure provides a method for reducing memory traffic in a distributed memory system. The method may include storing a presence vector in a directory of a memory slice, said presence vector indicating whether a line in local memory has been cached. The method may further include protecting said memory slice from cache coherency violations via a home agent configured to transmit and receive data from said memory slice, said home agent configured to store a copy of said presence vector. The method may also include receiving a request for a block of data from at least one processing node at said home agent and comparing said presence vector with said copy of said presence vector stored in said home agent. The method may additionally include eliminating a write update operation between said home agent and said directory if said presence vector and said copy are equivalent. Of course, many alternatives, variations and modifications are possible without departing from this embodiment.
摘要翻译：本公开提供一种用于减少分布式存储器系统中的存储器业务的方法。该方法可以包括将存在向量存储在存储器片的目录中，所述存在向量指示本地存储器中的行是否已被缓存。所述方法还可以包括：经由被配置为从所述存储器片传送和接收数据的归属代理来保护所述存储器切片免受高速缓存一致性冲突，所述归属代理被配置为存储所述存在向量的副本。该方法还可以包括从所述归属代理处的至少一个处理节点接收对数据块的请求，并将所述存在向量与存储在所述归属代理中的所述存在向量的所述副本进行比较。该方法可以另外包括：如果所述存在向量和所述副本是相同的，则消除所述归属代理和所述目录之间的写入更新操作。当然，在不脱离本实施例的情况下，可以进行许多替代，变化和修改。

3. 发明申请

US20090276581A1 METHOD, SYSTEM AND APPARATUS FOR REDUCING MEMORY TRAFFIC IN A DISTRIBUTED MEMORY SYSTEM 有权
标题翻译：用于减少分布式存储器系统中的存储器交通的方法，系统和装置
公开(公告)号：US20090276581A1
公开(公告)日：2009-11-05
申请号：US12113268
申请日：2008-05-01
申请人： Adrian Moga , Rajat Agarwal , Malcolm Mandviwalla
发明人： Adrian Moga , Rajat Agarwal , Malcolm Mandviwalla
IPC分类号： G06F12/08
CPC分类号： G06F12/082 , G06F2212/1041
摘要： The present disclosure provides a method for reducing memory traffic in a distributed memory system. The method may include storing a presence vector in a directory of a memory slice, said presence vector indicating whether a line in local memory has been cached. The method may further include protecting said memory slice from cache coherency violations via a home agent configured to transmit and receive data from said memory slice, said home agent configured to store a copy of said presence vector. The method may also include receiving a request for a block of data from at least one processing node at said home agent and comparing said presence vector with said copy of said presence vector stored in said home agent. The method may additionally include eliminating a write update operation between said home agent and said directory if said presence vector and said copy are equivalent. Of course, many alternatives, variations and modifications are possible without departing from this embodiment.
摘要翻译：本公开提供了一种用于减少分布式存储器系统中的存储器业务的方法。该方法可以包括将存在向量存储在存储器片的目录中，所述存在向量指示本地存储器中的行是否已被缓存。所述方法还可以包括：经由被配置为从所述存储器片传送和接收数据的归属代理来保护所述存储器切片免受高速缓存一致性冲突，所述归属代理被配置为存储所述存在向量的副本。该方法还可以包括从所述归属代理处的至少一个处理节点接收对数据块的请求，并将所述存在向量与存储在所述归属代理中的所述存在向量的所述副本进行比较。该方法可以另外包括：如果所述存在向量和所述副本是相同的，则消除所述归属代理和所述目录之间的写入更新操作。当然，在不脱离本实施例的情况下，可以进行许多替代，变化和修改。

4. 发明申请

US20130185522A1 ALLOCATION AND WRITE POLICY FOR A GLUELESS AREA-EFFICIENT DIRECTORY CACHE FOR HOTLY CONTESTED CACHE LINES 有权
标题翻译：用于高速缓存高速缓存线路的无障碍区域高效目录的分配和写入策略
公开(公告)号：US20130185522A1
公开(公告)日：2013-07-18
申请号：US13786305
申请日：2013-03-05
申请人： ADRIAN C. MOGA , MALCOLM MANDVIWALLA , VEDARAMAN GEETHA , HERBERT H. HUM
发明人： ADRIAN C. MOGA , MALCOLM MANDVIWALLA , VEDARAMAN GEETHA , HERBERT H. HUM
IPC分类号： G06F12/08
CPC分类号： G06F12/0831 , G06F12/082 , G06F2212/2542
摘要： Methods and apparatus relating to allocation and/or write policy for a glueless area-efficient directory cache for hotly contested cache lines are described. In one embodiment, a directory cache stores data corresponding to a caching status of a cache line. The caching status of the cache line is stored for each of a plurality of caching agents in the system. An write-on-allocate policy is used for the directory cache by using a special state (e.g., snoop-all state) that indicates one or more snoops are to be broadcasted to all agents in the system. Other embodiments are also disclosed.
摘要翻译：描述与用于热挑战的高速缓存行的无胶带区域高效目录高速缓存的分配和/或写入策略有关的方法和装置。在一个实施例中，目录高速缓存存储与高速缓存行的高速缓存状态对应的数据。为系统中的多个缓存代理中的每一个存储缓存行的高速缓存状态。通过使用指示要广播到系统中的所有代理的一个或多个窥探的特殊状态（例如，窥探全状态），对目录高速缓存使用写入分配策略。还公开了其他实施例。

5. 发明授权

US07962694B2 Partial way hint line replacement algorithm for a snoop filter 有权
标题翻译：用于窥探过滤器的部分方式提示行替换算法
公开(公告)号：US07962694B2
公开(公告)日：2011-06-14
申请号：US11639118
申请日：2006-12-14
申请人： Meenakshisundaram R. Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
发明人： Meenakshisundaram R. Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
IPC分类号： G06F12/16
CPC分类号： G06F12/0831 , G06F12/082
摘要： In an embodiment, a method is provided. The method of this embodiment provides receiving a request for data from a processor of a plurality of processors, determining a cache entry location based, at least in part, on the request, storing the data in a cache corresponding to the processor at the cache entry location, and storing a coherency record corresponding to the data in a snoop filter in accordance with one of the following, if there is a cache miss: at the cache entry location of a corresponding affinity in the snoop filter if the cache entry location is found in the corresponding affinity, or at a derived cache entry location of the corresponding affinity if the cache entry location is not found in the corresponding affinity.
摘要翻译：在一个实施例中，提供了一种方法。该实施例的方法提供从多个处理器的处理器接收对数据的请求，至少部分地基于请求确定高速缓存条目位置，将数据存储在与高速缓存条目处理器相对应的高速缓存中位置，并且如果存在高速缓存未命中，则根据以下之一存储与窥探过滤器中的数据相对应的一致性记录：如果发现高速缓存入口位置，则在窥探过滤器中的相应关联性的高速缓存入口位置处如果在相应的亲和度中没有找到高速缓存入口位置，则以相应的关联性，或在相应的亲和度的派生高速缓存入口位置。

6. 发明申请

US20070233966A1 Partial way hint line replacement algorithm for a snoop filter 有权
标题翻译：用于窥探过滤器的部分方式提示行替换算法
公开(公告)号：US20070233966A1
公开(公告)日：2007-10-04
申请号：US11639118
申请日：2006-12-14
申请人： Sundaram Chinthanmani , Kal Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
发明人： Sundaram Chinthanmani , Kal Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
IPC分类号： G06F13/00
CPC分类号： G06F12/0831 , G06F12/082
摘要： In an embodiment, a method is provided. The method of this embodiment provides receiving a request for data from a processor of a plurality of processors, determining a cache entry location based, at least in part, on the request, storing the data in a cache corresponding to the processor at the cache entry location, and storing a coherency record corresponding to the data in a snoop filter in accordance with one of the following, if there is a cache miss: at the cache entry location of a corresponding affinity in the snoop filter if the cache entry location is found in the corresponding affinity, or at a derived cache entry location of the corresponding affinity if the cache entry location is not found in the corresponding affinity.
摘要翻译：在一个实施例中，提供了一种方法。该实施例的方法提供从多个处理器的处理器接收对数据的请求，至少部分地基于请求确定高速缓存条目位置，将数据存储在与高速缓存条目处理器相对应的高速缓存中位置，并且如果存在高速缓存未命中，则根据以下之一存储与窥探过滤器中的数据相对应的一致性记录：如果发现高速缓存入口位置，则在窥探过滤器中的相应关联性的高速缓存入口位置处如果在相应的亲和度中没有找到高速缓存入口位置，则以相应的关联性，或在相应的亲和度的派生高速缓存入口位置。

7. 发明授权

US08055851B2 Line swapping scheme to reduce back invalidations in a snoop filter 有权
标题翻译：线路交换方案，以减少在窥探过滤器中的无效
公开(公告)号：US08055851B2
公开(公告)日：2011-11-08
申请号：US11638851
申请日：2006-12-14
申请人： Meenakshisundaram R. Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
发明人： Meenakshisundaram R. Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
IPC分类号： G06F13/14
CPC分类号： G06F12/0831 , G06F12/082
摘要： In an embodiment, a method is provided. The method of this embodiment provides receiving a request for data from a processor of a plurality of processors, determining a cache entry location based, at least in part, on the request, storing the data in a cache corresponding to the processor at the cache entry location, and storing a coherency record corresponding to the data in an affinity corresponding to the cache.
摘要翻译：在一个实施例中，提供了一种方法。该实施例的方法提供从多个处理器的处理器接收对数据的请求，至少部分地基于请求确定高速缓存条目位置，将数据存储在与高速缓存条目处理器相对应的高速缓存中并且以与所述高速缓存相对应的亲和度存储对应于所述数据的一致性记录。

8. 发明申请

US20080147986A1 Line swapping scheme to reduce back invalidations in a snoop filter 有权
标题翻译：线路交换方案，以减少在窥探过滤器中的无效
公开(公告)号：US20080147986A1
公开(公告)日：2008-06-19
申请号：US11638851
申请日：2006-12-14
申请人： Sundaram Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
发明人： Sundaram Chinthamani , Kai Cheng , Malcolm Mandviwalla , Bahaa Fahim , Keith R. Pflederer
IPC分类号： G06F12/00
CPC分类号： G06F12/0831 , G06F12/082
摘要： In an embodiment, a method is provided. The method of this embodiment provides receiving a request for data from a processor of a plurality of processors, determining a cache entry location based, at least in part, on the request, storing the data in a cache corresponding to the processor at the cache entry location, and storing a coherency record corresponding to the data in an affinity corresponding to the cache.
摘要翻译：在一个实施例中，提供了一种方法。该实施例的方法提供从多个处理器的处理器接收对数据的请求，至少部分地基于请求确定高速缓存条目位置，将数据存储在与高速缓存条目处理器相对应的高速缓存中并且以与所述高速缓存相对应的亲和度存储对应于所述数据的一致性记录。

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