公开(公告)号：US09864538B1

公开(公告)日：2018-01-09

申请号：US14750926

申请日：2015-06-25

Applicant: Amazon Technologies, Inc.

Inventor： Robert Michael Johnson ,  Mark Bradley Davis ,  Norbert Paul Kusters ,  Marc Stephen Olson ,  Marc John Brooker

IPC: G06F12/00 ,  G06F3/06 ,  G06F9/50

CPC classification number: G06F3/0626 ,  G06F3/0605 ,  G06F3/0631 ,  G06F3/0665 ,  G06F3/0689 ,  G06F9/5077

Abstract: Server computers often include one or more input/output (I/O) devices for communicating with a network or directly attached storage device. Data transferred between these devices may include blocks of data with a common often repeated and identifiable data pattern. Transfer and storage of data of this nature may be optimized by transferring primarily blocks of data that are not of the pre-determined data pattern. An indicator may be transferred and stored with transferred data that has been reduced in size in this manner.

公开(公告)号：US09817786B1

公开(公告)日：2017-11-14

申请号：US14752294

申请日：2015-06-26

Applicant: Amazon Technologies, Inc.

Inventor： Asif Khan ,  Thomas A. Volpe ,  Marc John Brooker ,  Marc Stephen Olson ,  Norbert Paul Kusters ,  Mark Bradley Davis ,  Robert Michael Johnson

IPC: G06F12/02 ,  G06F13/42 ,  G11C15/00 ,  G06F11/10

CPC classification number: G06F13/4221 ,  G06F11/1064 ,  G11C15/00

Abstract: Server computers often include one or more input/output (I/O) adapter devices for communicating with a network or directly attached storage device. The data transfer latency for request can be reduced by utilizing ingress data placement logic to bypass the processor of the I/O adapter device. For example, host memory descriptors can be stored in a content addressable memory unit of the I/O adapter device to facilitate placement of requested data.

公开(公告)号：US20160182398A1

公开(公告)日：2016-06-23

申请号：US14578004

申请日：2014-12-19

Applicant: Amazon Technologies, Inc.

Inventor： Mark Bradley Davis ,  David James Borland

IPC: H04L12/911 ,  G06F13/40 ,  G06F12/08 ,  H04L29/08

Abstract: Embodiments can provide additional computing resources at minimal and incremental cost by providing instances of one or more server compute subsystems on a system-on-chip. The system-on-chip can include multiple compute subsystems where each compute subsystem can include dedicated processing and memory resources. The system-on-chip can also include a management compute subsystem that can manage the processing and memory resources for each subsystem.

Abstract translation: 实施例可以通过在片上系统上提供一个或多个服务器计算子系统的实例来以最小和增加的成本提供额外的计算资源。 片上系统可以包括多个计算子系统，其中每个计算子系统可以包括专用处理和存储器资源。 片上系统还可以包括可管理每个子系统的处理和存储资源的管理计算子系统。