公开(公告)号：US20190042298A1

公开(公告)日：2019-02-07

申请号：US16144388

申请日：2018-09-27

Applicant: Intel Corporation

Inventor： Ranganath Sunku ,  Dinesh Kumar ,  Irene Liew ,  Kavindya Deegala ,  Sravanthi Tangeda

IPC: G06F9/455 ,  G06F9/50 ,  H04L12/931

Abstract: Technologies for adaptive platform resource management include a compute node to manage a processor core mapping scheme between virtual machines (VMs) and a virtual switch of the compute node via a set of virtual ports. The virtual switch is also coupled to a network interface controller (NIC) of the compute node via another set of virtual ports. Each of the VMs is configured to either provide ingress or egress to the NIC or provide ingress/egress across the VMs, via the virtual ports. The virtual ports for providing ingress or egress to the NIC are pinned to a same processor core of a processor of the compute node, and each of the virtual ports for providing ingress and/or egress across the VMs are pinned to a respective processor core of the processor such that data is transferred across VMs by coupled virtual ports that are pinned to the same processor core.

公开(公告)号：US09462084B2

公开(公告)日：2016-10-04

申请号：US14580826

申请日：2014-12-23

Applicant: Intel Corporation

Inventor： Patrick Connor ,  Ira Weiny ,  Iosif Gasparakis ,  Alexander W. Min ,  Andrew J. Herdrich ,  Dinesh Kumar ,  Tsung-Yuan C. Tai ,  Brian J. Skerry

IPC: H04L12/24 ,  H04L29/06

CPC classification number: G06F9/45558 ,  G06F2009/45595

Abstract: Technologies for identifying service functions that may be performed in parallel in a service function chain include a computing device for running one or more virtual machines for each of a plurality of service functions based on a preferred service function chain being selected. To identify which service functions may be performed in parallel, the computing device may determine which service functions are not required to be performed on a critical path of the service function chain and/or which service functions are not required to be performed in real-time. Additionally, selecting the preferred service function chain may be based on selection criteria.

Abstract translation: 用于识别可以在服务功能链中并行执行的服务功能的技术包括用于基于所选择的优选服务功能链来运行用于多个服务功能中的每一个的一个或多个虚拟机的计算设备。 为了识别可以并行执行哪些服务功能，计算设备可以确定不需要在服务功能链的关键路径上执行哪些服务功能和/或不需要实时执行哪些服务功能 。 此外，选择优选服务功能链可以基于选择标准。