Cloud computing architecture for managing hardware resources on network elements

• US 9,798,552 B2
• Filed: 07/15/2015
• Issued: 10/24/2017
• Est. Priority Date: 07/15/2014
• Status: Active Grant

• Current Assignee: Futurewei Technologies Incorporated (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Futurewei Technologies Incorporated (Huawei Investment & Holding Co., Ltd.)
• Inventors: Wei, An, Shen, Kangkang
• Primary Examiner(s): Rehman, Mohammed
• Application Number:
  US14/800,401
• Publication Number:
  US 20160019069A1
• Time in Patent Office: 832 Days
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 9/4401 Bootstrapping security arrangements therefor G06F21/57
• CPC Class Codes: G06F 9/4408 Boot device selection
• CPC Class Codes: G06F 9/4416 Network booting; Remote initial program loading [RIPL]
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06Q 30/018 Certifying business or products

Abstract

A network element (NE) comprising a receiver configured to couple to a cloud network; and a multi-core central processing unit (CPU) coupled to the receiver and configured to receive a first partition configuration from an orchestration element, partition a plurality of processor cores into a plurality of processor core partitions according to the first partition configuration, and initiate a plurality of virtual basic input/output systems (vBIOSs) such that each vBIOS manages a processor core partition.

Claims

1. A network element (NE) comprising:

a receiver configured to couple to a cloud network; and

a multi-core central processing unit (CPU) coupled to the receiver and configured to;

receive a first partition configuration from an orchestration element;

partition a plurality of processor cores into a plurality of processor core partitions according to the first partition configuration;

initiate a plurality of virtual basic input/output systems (vBIOSs) generated by the orchestration element to form a plurality of computing subsystems within the NE, wherein each vBIOS manages a processor core partition, wherein at least one vBIOS is configured to run an application without booting a corresponding operating system (OS) to manage the application; and

execute a distributed application across the plurality of computing subsystems, wherein the distributed application transmits hardware resource requests of applications executing in the plurality of computing subsystems to the orchestration element.

2. The NE of claim 1, wherein at least one vBIOS is distributed over a plurality of NE chassis such that the at least one vBIOS manages both a first processor core in a corresponding processor core partition on the NE and a second processor core in a corresponding remote processor core partition in a remote NE chassis coupled to the NE via the receiver.

3. The NE of claim 1, wherein the CPU comprises a physical BIOS, and wherein the physical BIOS is configured to partition the processor cores and initiate the plurality of vBIOSs based on the first partition configuration from the orchestration element.

4. The NE of claim 3, wherein the CPU is further configured to receive a second partition configuration from the orchestration element, and wherein the physical BIOS is further configured to unload at least one of the vBIOS without restarting the CPU based on the second partition configuration.

5. The NE of claim 4, wherein the physical BIOS is further configured to:

repartition at least one of the plurality of processor core partitions according to the second partition configuration; and

initiate a vBIOS in each repartitioned processor core partition.

6. The NE of claim 5, wherein at least one of the vBIOSs continues to operate on a corresponding processor core partition during repartitioning by the physical BIOS.

7. The NE of claim 1, wherein the receiver is configured to receive the first partition configuration from the orchestration element, wherein the CPU receives the first partition configuration via the receiver, and wherein the NE does not comprise the orchestration element.

8. The NE of claim 1, wherein the orchestration element operates at least in part on the NE and is distributed across the cloud network coupled to the NE via the receiver.

9. A method implemented in an orchestration element, the method comprising:

launching, via a processor in the orchestration element, a transient operating system environment;

determining, via the processor, a first cloud computing configuration comprising a plurality of computing subsystems formed within a plurality of network elements (NEs) in a cloud;

communicating, via a transmitter in the orchestration element, with the plurality of network elements (NEs) in the cloud to discover available hardware resources on the NEs;

communicating, via the transmitter, with the NEs to launch a plurality of physical basic input/output system (pBIOSs) on the NEs such that the pBIOSs manage the available hardware resources to implement the first cloud computing configuration including implementing a distributed application across the plurality of computing subsystems, wherein the distributed application transmits hardware resource requests of applications executing in the plurality of computing subsystems to the orchestration element; and

generating, via the processor, a plurality of virtual BIOSs (vBIOSs) such that each vBIOS corresponds to available hardware resources in a corresponding resource partition, wherein at least one vBIOS is configured to run an application without booting a corresponding operating system (OS) to manage the application.

10. The method of claim 9, further comprising:

communicating, via the transmitter, with a first of the pBIOSs on a first of the NEs to partition available hardware resources on the first NE into a plurality of resource partitions according to the first cloud computing configuration; and

communicating, via the transmitter, with the first pBIOS to initiate the plurality of vBIOSs such that each vBIOS manages the available hardware resources in the corresponding resource partition.

11. The method of claim 10, further comprising:

communicating, via the transmitter, with a second of the pBIOSs on a second of the NEs to partition available hardware resources on the second NE into a second plurality of resource partitions according to the first cloud computing configuration; and

communicating, via the transmitter, with the second pBIOS to initiate a first of the vBIOSs such that the first vBIOS manages available hardware resources on both the first NE and the second NE.

12. The method of claim 10, wherein at least one of the vBIOS is configured to operate a virtual machine (VM) without booting a corresponding operating system (OS).

13. The method of claim 10, further comprising:

determining, via the processor, a second cloud computing configuration which is different from the first cloud configuration; and

communicating, via the transmitter, with the first pBIOS to unload at least some of the vBIOS and repartition the first NE available resources according to the second cloud computing configuration without restarting the first NE.

14. The method of claim 13, wherein at least some of the vBIOS continue to operate during repartitioning.

15. A method implemented in a network element (NE), the method comprising:

receiving, via a port in the NE, a partition configuration from an orchestration element;

partitioning, by a processor in the NE, locally available hardware resources into a plurality of partitions according to the partition configuration;

initiating, by the processor, a plurality of virtual basic input/output systems (vBIOSs) generated by the orchestration element such that each vBIOS manages all hardware resources assigned to an associated partition to form a plurality of computing subsystems within the NE, wherein at least one vBIOS in the plurality of vBIOSs is configured to run an application without booting a corresponding operating system (OS) to manage the application; and

executing, by the processor, a distributed application across the plurality of computing subsystems, wherein the distributed application transmits hardware resource requests of applications executing in the plurality of computing subsystems to the orchestration element.

16. The method of claim 15, wherein at least one of the vBIOS is configured to boot a virtual machine (VM) without booting a corresponding operating system (OS).

17. The method of claim 16, wherein the at least one vBIOS is configured to boot the VM without booting a local virtual machine manager (VMM).

18. The method of claim 15, wherein at least one vBIOS is distributed over a plurality of NE chassis such that the at least one vBIOS manages both hardware resources in a corresponding partition on the NE and hardware resources in a corresponding remote partition in a remote NE chassis.

19. The method of claim 15, further comprising repartitioning, via the processor, the locally available hardware resources based on communications from the orchestration element, wherein at least one of the vBIOS remains operational during repartitioning of the locally available hardware resources.