会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 23. 发明申请
    • Pre-coherence channel
    • 前相干频道
    • US20070055828A1
    • 2007-03-08
    • US11221616
    • 2005-09-07
    • Benjamin Tsien
    • Benjamin Tsien
    • G06F13/28
    • G06F12/0831
    • A cache architecture to increase communication throughput and reduce stalls due to coherence protocol dependencies. More particularly, embodiments of the invention include multiple cache agents that each communication with the same protocol agent. In one embodiment, a pre-coherence channel couples the cache agents to the protocol agent to enable the protocol agent to receive events corresponding to cache operations from the cache agents to maintain ordering with respect to the cache operation events.
    • 缓存体系结构,用于增加通信吞吐量并减少由于一致性协议依赖关系造成的失速。 更具体地,本发明的实施例包括与相同协议代理进行通信的多个高速缓存代理。 在一个实施例中,预相干信道将高速缓存代理耦合到协议代理,以使得协议代理能够接收来自高速缓存代理的与高速缓存操作相对应的事件,以保持关于高速缓存操作事件的排序。
    • 26. 发明申请
    • PRE-COHERENCE CHANNEL
    • 预先相干通道
    • US20100262788A1
    • 2010-10-14
    • US12822599
    • 2010-06-24
    • Benjamin Tsien
    • Benjamin Tsien
    • G06F12/08G06F5/06
    • G06F12/0831
    • A cache architecture to increase communication throughput and reduce stalls due to coherence protocol dependencies. More particularly, embodiments of the invention include multiple cache agents that each communication with the same protocol agent. In one embodiment, a pre-coherence channel couples the cache agents to the protocol agent to enable the protocol agent to receive events corresponding to cache operations from the cache agents to maintain ordering with respect to the cache operation events.
    • 缓存体系结构,用于增加通信吞吐量并减少由于一致性协议依赖关系造成的失速。 更具体地,本发明的实施例包括与相同协议代理进行通信的多个高速缓存代理。 在一个实施例中,预相干信道将高速缓存代理耦合到协议代理,以使得协议代理能够接收来自高速缓存代理的与高速缓存操作相对应的事件,以保持关于高速缓存操作事件的排序。
    • 27. 发明授权
    • Remote hit predictor
    • 远程命中预测器
    • US07640401B2
    • 2009-12-29
    • US11691007
    • 2007-03-26
    • Benjamin Tsien
    • Benjamin Tsien
    • G06F12/00
    • G06F12/0826
    • In one embodiment, a first node comprises at least one memory request source and a node controller coupled to the memory request source. The node controller comprises a remote hit predictor configured to predict a second node to have a coherent copy of a block addressed by a memory request generated by the memory request source, and the node controller is configured to issued a speculative probe to the second node responsive to the prediction and to the memory request.
    • 在一个实施例中,第一节点包括耦合到存储器请求源的至少一个存储器请求源和节点控制器。 节点控制器包括远程命中预测器,其被配置为预测第二节点具有由存储器请求源生成的存储器请求寻址的块的相干副本,并且节点控制器被配置为向第二节点发出推测性探测器 到预测和存储器请求。