公开(公告)号：US10425275B2

公开(公告)日：2019-09-24

申请号：US14620365

申请日：2015-02-12

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James ,  Justin R. Unger

IPC: H04L12/24

Abstract: A management node of a server maintains a database of configuration parameters that stores the individual configuration parameters for each compute node. In response to a boot request at a compute node, a configuration parameter control module at the node intercepts locally targeted requests to load configuration parameters. The configuration parameter module sends the requests to the management node and in response the management node sends configuration parameters responsive to the requests. The configuration parameter module provides the configuration parameters to the compute node, thereby emulating a local boot memory for the compute node.

公开(公告)号：US20150333957A1

公开(公告)日：2015-11-19

申请号：US14620365

申请日：2015-02-12

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James ,  Justin R. Unger

IPC: H04L12/24

Abstract: A management node of a server maintains a database of configuration parameters that stores the individual configuration parameters for each compute node. In response to a boot request at a compute node, a configuration parameter control module at the node intercepts locally targeted requests to load configuration parameters. The configuration parameter module sends the requests to the management node and in response the management node sends configuration parameters responsive to the requests. The configuration parameter module provides the configuration parameters to the compute node, thereby emulating a local boot memory for the compute node.

Abstract translation: 服务器的管理节点维护存储每个计算节点的各个配置参数的配置参数的数据库。 响应于计算节点处的引导请求，节点处的配置参数控制模块拦截本地针对性的请求以加载配置参数。 配置参数模块将请求发送给管理节点，响应管理节点根据请求发送配置参数。 配置参数模块向计算节点提供配置参数，从而模拟计算节点的本地引导存储器。

公开(公告)号：US20150333956A1

公开(公告)日：2015-11-19

申请号：US14461614

申请日：2014-08-18

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James ,  Jean-Philippe Fricker

IPC: H04L12/24 ,  H04L29/08

CPC classification number: H04L41/0886 ,  G06F15/17387 ,  H04L41/0803 ,  H04L41/0806 ,  H04L41/16 ,  H04L45/48 ,  H04L67/10

Abstract: A cluster computer server is configured after a system reset or other configuration event. Each node of a fabric of the cluster compute server is employed, for purposes of configuration, as a cell in a cellular automaton, thereby obviating the need for a special configuration network to communicate configuration information from a central management unit. Instead, the nodes communicate configuration information using the same fabric interconnect that is used to communicate messages during normal execution of software services at the nodes.

Abstract translation: 在系统复位或其他配置事件后配置集群计算机服务器。 集群计算服务器的结构的每个节点被用于配置，作为细胞自动机中的小区，从而避免需要特殊配置网络来从中央管理单元传送配置信息。 相反，节点在节点处的软件服务的正常执行期间使用用于传送消息的相同的结构互连来传送配置信息。

公开(公告)号：US10158530B2

公开(公告)日：2018-12-18

申请号：US14461614

申请日：2014-08-18

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James ,  Jean-Philippe Fricker

IPC: G06F15/177 ,  H04L12/24 ,  H04L12/753 ,  G06F15/173

Abstract: A cluster computer server is configured after a system reset or other configuration event. Each node of a fabric of the cluster compute server is employed, for purposes of configuration, as a cell in a cellular automaton, thereby obviating the need for a special configuration network to communicate configuration information from a central management unit. Instead, the nodes communicate configuration information using the same fabric interconnect that is used to communicate messages during normal execution of software services at the nodes.

公开(公告)号：US09806908B2

公开(公告)日：2017-10-31

申请号：US14620342

申请日：2015-02-12

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James

IPC: H04L12/28 ,  H04L12/64

CPC classification number: H04L12/6418

Abstract: Each compute node of a cluster compute server generates and maintains route information for routing messages to other nodes of the server. Each compute node identifies the other nodes connected to a fabric interconnect and generates, based on a set of routing constraints, routes to each of the other nodes. Each compute node communicates messages to other nodes of the server via the generated routes. Because the routes are generated at each compute node the processing load to generate the routes is distributed among the compute nodes.

公开(公告)号：US20150334010A1

公开(公告)日：2015-11-19

申请号：US14620342

申请日：2015-02-12

Applicant: Advanced Micro Devices, Inc.

Inventor： Michael E. James

IPC: H04L12/721 ,  H04L12/933

CPC classification number: H04L12/6418

Abstract: Each compute node of a cluster compute server generates and maintains route information for routing messages to other nodes of the server. Each compute node identifies the other nodes connected to a fabric interconnect and generates, based on a set of routing constraints, routes to each of the other nodes. Each compute node communicates messages to other nodes of the server via the generated routes. Because the routes are generated at each compute node the processing load to generate the routes is distributed among the compute nodes.

Abstract translation: 集群计算服务器的每个计算节点生成并维护用于将消息路由到服务器的其他节点的路由信息​​。 每个计算节点标识连接到结构互连的其他节点，并基于一组路由约束生成到每个其他节点的路由。 每个计算节点通过生成的路由将消息传达到服务器的其他节点。 因为在每个计算节点处生成路由，所以在计算节点之间分配生成路由的处理负载。