公开(公告)号：US09185166B2

公开(公告)日：2015-11-10

申请号：US13407658

申请日：2012-02-28

Applicant: Casimer M. DeCusatis ,  Keshav G. Kamble ,  Abhijit P. Kumbhare ,  Renato J. Recio ,  Simarbir Singh

Inventor： Casimer M. DeCusatis ,  Keshav G. Kamble ,  Abhijit P. Kumbhare ,  Renato J. Recio ,  Simarbir Singh

IPC: H04L29/08 ,  H04L12/931 ,  H04L12/735 ,  H04L12/707

CPC classification number: H04L45/021 ,  H04L45/1283 ,  H04L45/24 ,  H04L49/357 ,  H04L67/1097

Abstract: According to one embodiment, a data center fabric network system includes a controller, a plurality of switches connected to the controller via a plurality of communication links, a source device connected to at least a first switch of the plurality of switches, a destination device connected to at least a second switch of the plurality of switches, and wherein the controller is adapted for: creating a topology of the data center fabric network system, determining paths between the plurality of switches, calculating one or more disjoint multi-paths from the source device to the destination device based on the topology, and writing the one or more disjoint multi-paths into a forwarding table of at least one of: the source device, the first switch, and a nearest shared switch to the source device.

Abstract translation: 根据一个实施例，数据中心结构网络系统包括控制器，经由多个通信链路连接到控制器的多个开关，连接到多个开关中的至少第一开关的源装置，连接的目的地装置 到所述多个交换机的至少第二交换机，并且其中所述控制器适于：创建所述数据中心结构网络系统的拓扑，确定所述多个交换机之间的路径，从所述源计算一个或多个不相交的多路径 将所述一个或多个不相交的多路径写入到所述源设备，所述源设备，所述第一交换机和最近的共享交换机中的至少一个的转发表中。

公开(公告)号：US08959310B2

公开(公告)日：2015-02-17

申请号：US13561010

申请日：2012-07-28

Applicant: Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

Inventor： Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

IPC: G06F12/00 ,  G06F12/02

CPC classification number: G06F12/0284 ,  Y02D10/13

Abstract: An approach is provided which a system selects a first virtual function from a plurality of virtual functions executing on a network adapter that includes a memory area. Next, the system allocates, in the memory area, a memory corresponding to the first virtual function. The system then stores one or more translation entries in the allocated memory partition, which are utilized to send data traversing through the first virtual function. As such, the system sends, utilizing one or more of the translation entries, the data packets from the network adapter to one or more destinations. In turn, the system dynamically resizes the memory partition based upon an amount of the memory partition that is utilized to store the one or more translation entries.

公开(公告)号：US08937940B2

公开(公告)日：2015-01-20

申请号：US13563390

申请日：2012-07-31

Applicant: Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

Inventor： Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

IPC: H04L12/28 ,  G06F9/455 ,  G06F15/167 ,  G06F13/28 ,  G06F12/02 ,  H04L12/931

CPC classification number: G06F9/45558 ,  H04L49/70

Abstract: An approach is provided in which a virtual function, which executes on a network adapter, receives a data packet from a first virtual machine. A translation entry is identified that corresponds to sending the data packet from the first virtual machine to a second virtual machine, and a determination is made as to whether an onboard memory partition assigned to the virtual function includes the identified translation. If the onboard memory location includes the translation entry, the data packet is sent to the destination virtual machine using the translation entry retrieved from the onboard memory partition. Otherwise, if the translation entry is not located in the onboard memory partition, the data packet is sent to the destination virtual machine using a translation entry retrieved from an off board memory location.

Abstract translation: 提供一种方法，其中在网络适配器上执行的虚拟功能从第一虚拟机接收数据分组。 识别对应于将数据分组从第一虚拟机发送到第二虚拟机的翻译条目，并且确定分配给虚拟功能的机载存储器分区是否包括所识别的翻译。 如果车载存储器位置包括转换条目，则使用从板载存储器分区检索的转换条目将数据分组发送到目标虚拟机。 否则，如果翻译条目不在板载内存分区中，则使用从脱机内存位置检索的转换条目将数据包发送到目标虚拟机。

公开(公告)号：US08645974B2

公开(公告)日：2014-02-04

申请号：US12111020

申请日：2008-04-28

Applicant: William J. Armstrong ,  Charles S. Graham ,  Sandy K. Kao ,  Kyle A. Lucke ,  Naresh Nayar ,  Michal Ostrowski ,  Renato J. Recio ,  Randal C. Swanberg

Inventor： William J. Armstrong ,  Charles S. Graham ,  Sandy K. Kao ,  Kyle A. Lucke ,  Naresh Nayar ,  Michal Ostrowski ,  Renato J. Recio ,  Randal C. Swanberg

IPC: G06F3/00 ,  G06F9/44 ,  G06F9/46 ,  G06F13/00 ,  G06F9/455

CPC classification number: G06F12/1475 ,  G06F9/45558 ,  G06F12/1491 ,  G06F2009/45579 ,  G06F2212/1056

Abstract: Multiple logical partitions are provided access to a self-virtualizing input/output device of a data processing system via multiple dedicated partition adjunct instances. Access is established by: interfacing each logical partition to one or more associated partition adjunct instances, each partition adjunct instance coupling its associated logical partition to one of a virtual function or a queue pair of the self-virtualizing input/output device, and each partition adjunct instance being a separate dispatchable state and being created employing virtual address space donated from the respective logical partition or a hypervisor of the data processing system, and each partition adjunct instance including a device driver for the virtual function or queue pair of the self-virtualizing input/output device; and providing each logical partition with at least one virtual input/output which is interfaced through the logical partition's respective partition adjunct instance(s) to a virtual function or queue pair of the self-virtualizing input/output device.

Abstract translation: 提供了多个逻辑分区，通过多个专用分区附件实例访问数据处理系统的自身虚拟化输入/输出设备。 访问是通过以下方式建立的：将每个逻辑分区连接到一个或多个相关联的分区附件实例，每个分区附件实例将其相关联的逻辑分区耦合到自虚拟输入/输出设备的虚拟功能或队列对之一，以及每个分区 辅助实例是单独的可分派状态，并且被使用从相应的逻辑分区或数据处理系统的管理程序捐赠的虚拟地址空间被创建，并且每个分区附件实例包括用于虚拟功能的虚拟功能或队列对的自身虚拟化 输入/输出设备; 以及向每个逻辑分区提供至少一个虚拟输入/输出，所述至少一个虚拟输入/输出通过所述逻辑分区的相应分区附件实例与所述自虚拟化输入/输出设备的虚拟功能或队列对连接。

公开(公告)号：US08582462B2

公开(公告)日：2013-11-12

申请号：US13006919

申请日：2011-01-14

Applicant: Rakesh Sharma ,  Renato J. Recio

Inventor： Rakesh Sharma ,  Renato J. Recio

IPC: H04L12/26

CPC classification number: G06F9/5077 ,  H04L45/66 ,  H04L69/14 ,  H04L69/32

Abstract: A physical host executes a virtual machine monitor (VMM) that instantiates a plurality of virtual machines (VMs). The VMM supports processing of a virtual link manager (VLM) that deploys and configures a plurality of Layer 2 virtual links sharing bandwidth of a Layer 2 physical link between the physical host and an access switch. The VMM communicates parameters of the plurality of virtual links with the access switch.

Abstract translation: 物理主机执行实例化多个虚拟机（VM）的虚拟机监视器（VMM）。 VMM支持处理虚拟链路管理器（VLM），该虚拟链路管理器（VLM）部署和配置共享物理主机和接入交换机之间的第2层物理链路的带宽的多个第2层虚拟链路。 VMM将该多个虚拟链路的参数与访问交换机进行通信。

公开(公告)号：US20130091261A1

公开(公告)日：2013-04-11

申请号：US13253338

申请日：2011-10-05

Applicant: Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

Inventor： Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

IPC: G06F15/173

CPC classification number: H04L41/5096 ,  H04L63/20

Abstract: An approach is provided in which a computer system selects a virtual domain from multiple virtual domains, which are each overlayed onto a physical network and are independent of physical topology constraints of the physical network. The computer system selects, from the selected virtual domain, a first virtual group that includes one or more first virtual network endpoints. Next, the computer system selects, from the selected virtual domain, a second virtual group that includes one or more second virtual network endpoints. In turn, the computer system creates a logical link policy that includes one or more actions corresponding to sending data between the first virtual group and the second virtual group.

Abstract translation: 提供了一种方法，其中计算机系统从多个虚拟域中选择虚拟域，每个虚拟域各自覆盖在物理网络上，并且与物理网络的物理拓扑约束无关。 计算机系统从所选择的虚拟域中选择包括一个或多个第一虚拟网络端点的第一虚拟组。 接下来，计算机系统从所选择的虚拟域中选择包括一个或多个第二虚拟网络端点的第二虚拟组。 反过来，计算机系统创建逻辑链路策略，该逻辑链路策略包括与在第一虚拟组和第二虚拟组之间发送数据相对应的一个或多个动作。

公开(公告)号：US20130042086A1

公开(公告)日：2013-02-14

申请号：US13561010

申请日：2012-07-28

Applicant: Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

Inventor： Omar Cardona ,  Vinit Jain ,  Jayakrishna Kidambi ,  Renato J. Recio

IPC: G06F12/02

CPC classification number: G06F12/0284 ,  Y02D10/13

Abstract: An approach is provided which a system selects a first virtual function from a plurality of virtual functions executing on a network adapter that includes a memory area. Next, the system allocates, in the memory area, a memory corresponding to the first virtual function. The system then stores one or more translation entries in the allocated memory partition, which are utilized to send data traversing through the first virtual function. As such, the system sends, utilizing one or more of the translation entries, the data packets from the network adapter to one or more destinations. In turn, the system dynamically resizes the memory partition based upon an amount of the memory partition that is utilized to store the one or more translation entries.

公开(公告)号：US08370447B2

公开(公告)日：2013-02-05

申请号：US13535560

申请日：2012-06-28

Applicant: Alan F. Benner ,  Michael A. Ko ,  Gregory F. Pfister ,  Renato J. Recio ,  Jacobo A. Vargas

Inventor： Alan F. Benner ,  Michael A. Ko ,  Gregory F. Pfister ,  Renato J. Recio ,  Jacobo A. Vargas

IPC: G06F15/167

CPC classification number: H04L67/1097 ,  G06F12/1466

Abstract: A system and method for providing a memory region/memory window (MR/MW) access notification on a system area network are provided. Whenever a previously allocated MR/MW is accessed, such as via a remote direct memory access (RDMA) read/write operation, a notification of the access is generated and written to a queue data structure associated with the MR/MW. In one illustrative embodiment, this queue data structure may be a MR/MW event queue (EQ) data structure that is created and used for all consumer processes and all MR/MWs. In other illustrative embodiments, the EQ is associated with a protection domain. In yet another illustrative embodiment, an event record may be posted to an asynchronous event handler in response to the accessing of the MR/MW. In another illustrative embodiment, a previously posted queue element may be used to generate a completion queue element in response to the accessing of the MR/MW.

Abstract translation: 提供了一种用于在系统区域网络上提供存储器区域/存储器窗口（MR / MW）访问通知的系统和方法。 无论何时访问先前分配的MR / MW，例如通过远程直接存储器访问（RDMA）读/写操作，生成访问通知并将其写入与MR / MW相关联的队列数据结构。 在一个说明性实施例中，该队列数据结构可以是创建并用于所有消费者进程和所有MR / MW的MR / MW事件队列（EQ）数据结构。 在其他说明性实施例中，EQ与保护域相关联。 在另一说明性实施例中，可以响应于MR / MW的访问而将事件记录发布到异步事件处理程序。 在另一示例性实施例中，可以使用先前发布的队列元素来响应于MR / MW的访问来生成完成队列元素。

公开(公告)号：US20120290703A1

公开(公告)日：2012-11-15

申请号：US13556769

申请日：2012-07-24

Applicant: Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

Inventor： Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

IPC: G06F15/173

CPC classification number: H04L41/0893 ,  H04L63/20

Abstract: According to one embodiment of the present disclosure, an approach is provided in which a policy server receives a request for a policy from a requestor. The policy server identifies an initiating virtual machine; the initial virtual machine's corresponding virtual network; and a destination virtual machine. Next, a policy corresponding to sending data from the first virtual machine to the second virtual machine is selected. The policy includes one or more logical references to the virtual network and does not include a physical reference to a physical entity located on a physical network. In turn, a physical path translation corresponding to the selected policy is identified and sent to the requestor.

Abstract translation: 根据本公开的一个实施例，提供了一种方法，其中策略服务器从请求者接收对策略的请求。 策略服务器识别启动虚拟机; 初始虚拟机的对应虚拟网络; 和目标虚拟机。 接下来，选择对应于从第一虚拟机向第二虚拟机发送数据的策略。 策略包括对虚拟网络的一个或多个逻辑引用，并且不包括对位于物理网络上的物理实体的物理引用。 反过来，识别对应于所选策略的物理路径转换并将其发送到请求者。

公开(公告)号：US20120290695A1

公开(公告)日：2012-11-15

申请号：US13107434

申请日：2011-05-13

Applicant: Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

Inventor： Katherine Barabash ,  Rami Cohen ,  Vinit Jain ,  Renato J. Recio ,  Benny Rochwerger

IPC: G06F15/16

CPC classification number: H04L41/0893 ,  H04L63/20

Abstract: According to one embodiment of the present disclosure, an approach is provided in which a policy server receives a request for a policy from a requestor. The policy server identifies an initiating virtual machine; the initial virtual machine's corresponding virtual network; and a destination virtual machine. Next, a policy corresponding to sending data from the first virtual machine to the second virtual machine is selected. The policy includes one or more logical references to the virtual network and does not include a physical reference to a physical entity located on a physical network. In turn, a physical path translation corresponding to the selected policy is identified and sent to the requestor.