Real-time local and global datacenter network optimizations based on platform telemetry data

Patent metadata

Field	Value
Publication number	US-10860374-B2
Application number	US-201514866869-A
Country	US
Kind code	B2
Filing date	Sep 26, 2015
Priority date	Sep 26, 2015
Publication date	Dec 8, 2020
Grant date	Dec 8, 2020

Abstract

In one embodiment, a system comprises platform logic comprising a plurality of processor cores and resource allocation logic. The resource allocation logic may receive a processing request and direct the processing request to a processor core of the plurality of processor cores, wherein the processor core is selected based at least in part on telemetry data associated with the platform logic, the telemetry data indicating a topology of at least a portion of the platform logic.

First claim

What is claimed is:

1. A system comprising:
a platform comprising a plurality of processor cores; and
an input/output (I/O) device driver to:
receive telemetry comprising global network topology data of a datacenter, including available bandwidth between processor cores and other resources;
receive a processing request; and
direct the processing request to a local processor core of the plurality of processor cores, comprising sending an interrupt by the I/O device driver to the local processor core, wherein the local processor core is selected by the I/O device driver based at least in part on the global network topology data.

2. The system of claim 1 , the I/O device driver further to run benchmark tests on the platform to collect the global network topology data.

3. The system of claim 1 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the local processor core to a memory of the platform.

4. The system of claim 1 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the local processor core to another processor core of the plurality of processor cores.

5. The system of claim 1 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the local processor core to an I/O device of the platform.

6. The system of claim 1 , wherein the local processor core is selected further based at least in part on current bandwidths used by the plurality of processor cores.

7. The system of claim 6 , wherein the I/O device driver is to select the local processor core based on the local processor core having a highest expected performance for the processing request of the plurality of processor cores when an aggregate used bandwidth of the plurality of processor cores is below a particular threshold.

8. The system of claim 6 , wherein the I/O device driver is to select the local processor core based on an optimization of overall bandwidth of the plurality of processor cores when an aggregate used bandwidth of the plurality of processor cores is above a particular threshold.

9. The system of claim 1 , wherein the processing request comprises a request to instantiate an I/O device driver.

10. The system of claim 1 , wherein the processing request comprises a request to instantiate a virtual machine.

11. The system of claim 1 , wherein the processing request is associated with a service function chain or a virtual network function running on the platform.

12. The system of claim 1 , further comprising a manageability engine to collect additional global network topology data associated with the platform in a manner that is out-of-band with respect to the plurality of processor cores.

13. The system of claim 12 , wherein the manageability engine is further to send the additional global network topology data to a datacenter management platform that receives global network topology data from a plurality of distinct platforms, each platform comprising at least one processor core.

14. The system of claim 12 , wherein the manageability engine is further to run benchmark tests on the platform to collect the global network topology data indicating a topology of at least a portion of the platform.

15. The system of claim 1 , wherein a hypervisor or operating system executed by the platform is to run benchmark tests on the platform to collect the telemetry comprising global network topology data including available bandwidth between processor cores and other resources.

16. The system of claim 1 , further comprising a hypervisor comprising the I/O device driver, the I/O device driver of the hypervisor to select the local processor core based further on one or more performance requirements associated with the processing request.

17. A method comprising:
receiving, at an input/output device driver, a processing request;
selecting, by the input/output device driver, a local processor core of a plurality of processor cores based at least in part on telemetry comprising global network topology data of a datacenter and further comprising an indication of a topology of at least a portion of the plurality of processor cores, the topology including available bandwidth between processor cores and other resources; and
sending, by the input/output device driver, an interrupt signal associated with the processing request to the selected local processor core.

18. The method of claim 17 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the local processor core to another processor core of the plurality of processor cores.

19. At least one non-transitory machine readable storage medium having instructions stored thereon, the instructions when executed by a machine to cause the machine to:
receive, at an input/output device driver, a processing request;
select, by the input/output device driver, a local processor core of a plurality of processor cores based at least in part on telemetry comprising global network topology data of a datacenter, including available bandwidth between processor cores and other resources; and
send, by the input/output device driver, an interrupt signal associated with the processing request to the selected local processor core.

20. The medium of claim 19 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the selected local processor core to another processor core of the plurality of processor cores.

21. The medium of claim 19 , wherein the global network topology data comprises an indication of a bandwidth or a latency from the selected local processor core to an I/O device.

22. The medium of claim 19 , wherein the instructions when executed are further to cause the machine to run benchmark tests on the plurality of processor cores to collect the global network topology data.

23. The medium of claim 19 , wherein the processing request comprises a request to instantiate an I/O device driver.

Assignees

• Intel Corp

Inventors

• Hearn James Robert
• Connor Patrick
• Sood Kapil
• Dubai Scott P
• Herdrich Andrew J

Classifications

• G06F9/5027 Primary

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06F2209/502

  Proximity · CPC title

• Y02D10/00

  Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title

• G06F9/4411

  Configuring for operating with peripheral devices; Loading of device drivers · CPC title

• G06F9/4881

  Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues · CPC title