公开(公告)号：US09571421B1

公开(公告)日：2017-02-14

申请号：US14047894

申请日：2013-10-07

Applicant: Amazon Technologies, Inc.

Inventor： Jagwinder Singh Brar ,  Michael David Marr ,  Tyson J. Lamoreaux ,  Mark N. Kelly ,  Justin O. Pietsch

IPC: H04L12/933 ,  H04L12/947 ,  H04Q11/00

CPC classification number: H04L49/1515 ,  H04L49/15 ,  H04L49/25 ,  H04Q3/68 ,  H04Q11/0005 ,  H04Q2011/0058

Abstract: Efficient and highly-scalable network solutions are provided that each utilize deployment units based on Clos networks, but in an environment such as a data center of Internet Protocol-based network. Each of the deployment units can include multiple stages of devices, where connections between devices are only made between stages and the deployment units are highly connected. In some embodiments, the level of connectivity between two stages can be reduced, providing available connections to add edge switches and additional host connections while keeping the same number of between-tier connections. In some embodiments, where deployment units (or other network groups) can be used at different levels to connect other deployment units, the edges of the deployment units can be fused to reduce the number of devices per host connection.

Abstract translation: 提供了高效和高度可扩展的网络解决方案，其每个利用基于Clos网络的部署单元，但在诸如基于因特网协议的网络的数据中心之类的环境中。 每个部署单元可以包括多个设备阶段，其中设备之间的连接仅在阶段之间进行，部署单元高度连接。 在一些实施例中，可以减少两级之间的连接级别，为保留相同数量的层间连接提供可用于添加边缘交换机和附加主机连接的连接。 在一些实施例中，其中可以在不同级别使用部署单元（或其他网络组）来连接其他部署单元，可以融合部署单元的边缘以减少每个主机连接的设备数量。

公开(公告)号：US09547353B1

公开(公告)日：2017-01-17

申请号：US13622901

申请日：2012-09-19

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr ,  Matthew D. Klein

IPC: G06F12/00 ,  G06F1/32 ,  G06F9/30 ,  G06F9/48 ,  G06F9/50

CPC classification number: G06F1/3206 ,  G06F1/3203 ,  G06F1/3228 ,  G06F1/324 ,  G06F1/329 ,  G06F1/3296 ,  G06F9/30076 ,  G06F9/30083 ,  G06F9/4893 ,  G06F9/5094 ,  G06Q10/06 ,  G06Q50/06 ,  Y02D10/126 ,  Y02D10/172 ,  Y02D10/22 ,  Y02D10/24 ,  Y02P90/82

Abstract: Customers in a multi-tenant environment can obtain energy consumption information for a set of resources or other computing components used by those customers, including time-accurate accounting for various components of those resources utilized on behalf of the customer. A customer can also have the ability to specify how the resources are to be operated when used for the customer, in order to manage the amount of energy consumption. The accounting can be performed even when the resources are shared among multiple users or entities. Various hardware components or agents can be used to provide detailed energy consumption information for those components that is associated with a particular customer. The information can be used not only for accounting and monitoring purposes, but also to make dynamic adjustments based on various changes in usage, energy consumption, or other such factors.

Abstract translation: 多租户环境中的客户可以获得这些客户使用的一组资源或其他计算组件的能耗信息，包括对代表客户使用的资源的各种组件进行时间准确的计算。 客户还可以指定如何在为客户使用资源时进行操作，以便管理能源消耗量。 即使资源在多个用户或实体之间共享，也可以进行计费。 可以使用各种硬件组件或代理来为与特定客户相关联的组件提供详细的能量消耗信息。 该信息不仅可以用于会计和监控，还可以根据使用，能源消耗或其他因素的各种变化进行动态调整。

公开(公告)号：US20170013053A1

公开(公告)日：2017-01-12

申请号：US15276491

申请日：2016-09-26

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr

IPC: H04L29/08 ,  H04L12/803 ,  H04L29/06

CPC classification number: H04L67/1019 ,  H04L47/125 ,  H04L67/1095 ,  H04L67/42

Abstract: Host machines and other devices performing synchronized operations can be dispersed across multiple racks in a data center to provide additional buffer capacity and to reduce the likelihood of congestion. The level of dispersion can depend on factors such as the level of oversubscription, as it can be undesirable in a highly connected network to push excessive host traffic into the aggregation fabric. As oversubscription levels increase, the amount of dispersion can be reduced and two or more host machines can be clustered on a given rack, or otherwise connected through the same edge switch. By clustering a portion of the machines, some of the host traffic can be redirected by the respective edge switch without entering the aggregation fabric. When provisioning hosts for a customer, application, or synchronized operation, for example, the levels of clustering and dispersion can be balanced to minimize the likelihood for congestion throughout the network.

Abstract translation: 执行同步操作的主机和其他设备可以分散在数据中心中的多个机架上，以提供额外的缓冲器容量并减少拥塞的可能性。 分散级别可以取决于诸如超额预订级别的因素，因为在高度连接的网络中可能不希望将过多的主机业务推送到聚合结构中。 随着超额认购水平的增加，可以减少分散量，并且可以将两台或多台主机集群在给定的机架上，或者通过相同的边缘交换机连接。 通过对一部分机器进行聚类，一些主机流量可以由相应的边缘交换机重定向，而不进入聚合结构。 当为客户，应用程序或同步操作配置主机时，可以平衡集群和分散的级别，以最小化整个网络拥塞的可能性。

公开(公告)号：US20160337265A1

公开(公告)日：2016-11-17

申请号：US15222487

申请日：2016-07-28

Applicant: Amazon Technologies, Inc.

Inventor： Jagwinder Singh Brar ,  Michael David Marr ,  Tyson J. Lamoreaux ,  Mark N. Kelly ,  Justin O. Pietsch

IPC: H04L12/933 ,  H04L12/931

CPC classification number: H04L49/1515 ,  H04L49/15 ,  H04L49/356

Abstract: Efficient and highly-scalable network solutions are provided that each utilize deployment units based on Clos networks, but in an environment such as a data center of Internet Protocol-based network. Each of the deployment units can include multiple stages of devices, where connections between devices are only made between stages and the deployment units are highly connected. In some embodiments, the level of connectivity between two stages can be reduced, providing available connections to add edge switches and additional host connections while keeping the same number of between-tier connections. In some embodiments, where deployment units (or other network groups) can be used at different levels to connect other deployment units, the edges of the deployment units can be fused to reduce the number of devices per host connection.

Abstract translation: 提供了高效和高度可扩展的网络解决方案，其每个利用基于Clos网络的部署单元，但在诸如基于因特网协议的网络的数据中心之类的环境中。 每个部署单元可以包括多个设备阶段，其中设备之间的连接仅在阶段之间进行，部署单元高度连接。 在一些实施例中，可以减少两级之间的连接级别，为保留相同数量的层间连接提供可用于添加边缘交换机和附加主机连接的连接。 在一些实施例中，其中可以在不同级别使用部署单元（或其他网络组）来连接其他部署单元，可以融合部署单元的边缘以减少每个主机连接的设备数量。

公开(公告)号：US20160180099A1

公开(公告)日：2016-06-23

申请号：US15056826

申请日：2016-02-29

Applicant: Amazon Technologies, Inc.

Inventor： Nachiketh Rao Potlapally ,  Michael David Marr

IPC: G06F21/60 ,  H04L9/08 ,  H04L9/30

CPC classification number: G06F21/602 ,  G06F21/572 ,  H04L9/0838 ,  H04L9/30

Abstract: Approaches are described for updating code and/or instructions in one or more computing devices. In particular, various embodiments provide approaches for updating the microcode of one or more processors of a computing device without requiring a restart of the computing device and without disrupting the various components (e.g., applications, virtual machines, etc.) executing on the computing device. The microcode updates can be performed on host computing devices deployed in a resource center of a service provider (e.g., cloud computing service provider), where each host computing device may be executing a hypervisor hosting multiple guest virtual machines (or other guest applications) for the customers of the service provider.

Abstract translation: 描述了用于在一个或多个计算设备中更新代码和/或指令的方法。 具体地，各种实施例提供了用于更新计算设备的一个或多个处理器的微代码的方法，而不需要重新启动计算设备，而不会中断在计算设备上执行的各种组件（例如，应用，虚拟机等） 。 可以在部署在服务提供商（例如，云计算服务提供商）的资源中心的主机计算设备上执行微代码更新，其中每个主机计算设备可以执行托管多个客体虚拟机（或其他客户应用）的虚拟机管理程序，用于 服务提供商的客户。

公开(公告)号：US09372731B1

公开(公告)日：2016-06-21

申请号：US13706729

申请日：2012-12-06

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr ,  Anirudh Balachandra Aithal ,  Matthew David Klein

IPC: G06F9/46 ,  G06F9/50

CPC classification number: G06F9/5077 ,  G06F8/654 ,  G06F9/4401 ,  G06F9/45533

Abstract: Systems and methods are described for managing computing resources. In one embodiment, mappings between a plurality of parameters of an abstracted firmware framework to corresponding firmware settings of computing components are maintained. The mappings are determined based on predetermined associations between vendor-specific firmware settings and abstracted firmware settings that implement a standardized interface that is independent of the vendor-specific firmware settings. In response to receiving one of the plurality of parameters, the received parameter is translated to corresponding vendor-specific firmware settings based on the mappings.

Abstract translation: 描述了管理计算资源的系统和方法。 在一个实施例中，保持抽象固件框架的多个参数与计算组件的对应固件设置之间的映射。 基于供应商特定固件设置和实现独立于供应商特定固件设置的标准化接口的抽象固件设置之间的预定关联来确定映射。 响应于接收到多个参数中的一个，所接收的参数被转换为基于映射的相应的供应商特定的固件设置。

公开(公告)号：US20160170782A1

公开(公告)日：2016-06-16

申请号：US14968625

申请日：2015-12-14

Applicant: Amazon Technologies, Inc.

Inventor： Matthew D. Klein ,  Michael David Marr ,  Samuel J. McKelvie

IPC: G06F9/455

CPC classification number: G06F9/45533 ,  G06F9/45558 ,  G06F2009/45575 ,  G06F2009/45595

Abstract: The transmission of data on computer networks according to one or more policies is disclosed. A policy may specify, among other things, various parameters which are to be followed when transmitting initiating network traffic. Multiple network interfaces may be installed on a server to enable transmission of data from the single server according a number of discrete configuration settings implicated by the various policies. The multiple network interfaces may correspond to separate physical components, with each component configured independently to implement a feature of a policy. The multiple network interfaces may also correspond to a single physical component that exposes multiple network interfaces, both to the network and to the server on which it is installed.

Abstract translation: 公开了根据一个或多个策略在计算机网络上传输数据。 除了别的以外，策略可以指定在发送启动网络流量时要遵循的各种参数。 可以在服务器上安装多个网络接口，以便根据各种策略所牵涉到的多个离散配置设置来传输来自单个服务器的数据。 多个网络接口可以对应于单独的物理组件，其中每个组件被独立地配置以实现策略的特征。 多个网络接口也可以对应于暴露多个网络接口的单个​​物理组件，无论是网络还是安装在其上的服务器。

公开(公告)号：US09349010B2

公开(公告)日：2016-05-24

申请号：US14671933

申请日：2015-03-27

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr ,  Pradeep Vincent ,  Matthew T. Corddry ,  James R. Hamilton

IPC: G06F3/00 ,  G06F13/28 ,  G06F11/30 ,  G06F9/44 ,  G06F21/57 ,  G06F9/445 ,  H04L29/08

CPC classification number: G06F21/572 ,  G06F8/65 ,  G06F11/3003 ,  G06F11/3051 ,  G06F21/57 ,  G06F21/575 ,  G06F21/577 ,  G06F2201/865 ,  G06F2201/88 ,  G06F2221/033 ,  H04L67/10

Abstract: Attempts to update confirmation information or firmware for a hardware device can be monitored using a secure counter that is configured to monotonically adjust a current value of the secure counter for each update or update attempt. The value of the counter can be determined every time the validity of the firmware is confirmed, and this value can be stored to a secure location. At subsequent times, such as during a boot process, the actual value of the counter can be determined and compared with the expected value. If the values do not match, such that the firmware may be in an unexpected state, an action can be taken, such as to prevent access to, or isolate, the hardware until such time as the firmware can be validated or updated to an expected state.

Abstract translation: 可以使用安全计数器监视硬件设备的更新确认信息或固件的尝试，该计数器被配置为对每次更新或更新尝试单调地调整安全计数器的当前值。 每次确认固件的有效性时，可以确定计数器的值，并将该值存储到安全位置。 在随后的时间，例如在引导过程期间，可以确定计数器的实际值并将其与预期值进行比较。 如果值不匹配，使得固件可能处于意外状态，则可以采取措施，例如防止访问或隔离硬件，直到固件可以被验证或更新为预期的时间 州。

公开(公告)号：US09183092B1

公开(公告)日：2015-11-10

申请号：US13746228

申请日：2013-01-21

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr ,  Matthew D. Klein

IPC: G06F11/14 ,  G06F9/44

CPC classification number: G06F11/1417 ,  G06F9/4401

Abstract: A system and method for preventing dependency problems, such as deadlocks, within startup of computing service workflows, such as workflows that occur within computing assets that provide network-based computing services. The system and method creates a remedial workflow or action for the computing services to address deadlocks or other blocking conditions within the services which may occur should the underlying computing assets need to be restarted, rebooted or sequentially execute and reach a problematic operational state. The system and method will determine the reliance of each computing service upon the functionality of one or more other network-based computing services and structure the remedial workflow accordingly. Other aspects of the disclosure are described in the detailed description, figures, and claims.

Abstract translation: 用于在计算服务工作流的启动期间，例如在提供基于网络的计算服务的计算资产内发生的工作流，防止依赖问题（例如死锁）的系统和方法。 系统和方法为计算服务创建补救工作流或操作，以解决服务中的死锁或其他阻塞条件，如果底层计算资产需要重新启动，重新启动或顺序执行并达到有问题的操作状态，可能会发生。 系统和方法将确定每个计算服务对一个或多个其他基于网络的计算服务的功能的依赖，并相应地构建补救工作流。 在详细描述，附图和权利要求中描述了本公开的其它方面。

公开(公告)号：US09176752B1

公开(公告)日：2015-11-03

申请号：US13693946

申请日：2012-12-04

Applicant: Amazon Technologies, Inc.

Inventor： Michael David Marr ,  Nachiketh Rao Potlapally

IPC: G06F9/24 ,  G06F15/177 ,  G06F9/445

CPC classification number: G06F9/4812 ,  G06F8/65 ,  G06F8/656 ,  G06F9/4411 ,  G06F9/4418 ,  G06F9/44505 ,  H04L9/0877

Abstract: Techniques for using hardware-based mechanisms for updating computing resources are described herein. At a time after receiving a code update request, one or more hardware-supported system management capabilities of processors within a computing system are invoked at least to interrupt execution of currently running instructions. While the system management capabilities are active and instruction execution is suspended, programmatic routines are updated. After the updates are complete, instruction execution is resumed.

Abstract translation: 本文描述了使用基于硬件的机制来更新计算资源的技术。 在接收到代码更新请求之后，至少调用计算系统内的处理器的一个或多个硬件支持的系统管理能力来中断当前运行的指令的执行。 当系统管理功能处于活动状态并且指令执行被暂停时，程序化例程被更新。 更新完成后，指令执行恢复。