Parallel processing method supporting virtual core automatic scaling and apparatus therefor

• US 10,635,478 B2
• Filed: 11/22/2017
• Issued: 04/28/2020
• Est. Priority Date: 12/01/2016
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Choi, Kang Il, Lee, Bhum Cheol
• Primary Examiner(s): Nguyen, Van H
• Application Number:
  US15/821,259
• Publication Number:
  US 20180157520A1
• Time in Patent Office: 888 Days
• Field of Search: None
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 49/70 Virtual switches

Abstract

A parallel processing method for a traffic having an order, which supports virtual core automatic scaling in a virtual environment having at least one virtual machine (VM) and at least one virtual core queue for each VM, may comprise identifying a flow of the traffic according to attributes of the traffic; determining a virtual core queue to which the traffic is mapped according to the identified flow and queuing the traffic to the determined virtual core queue; and transferring the traffic from the determined virtual core queue to a VM corresponding to the determined virtual core queue through a virtual switch. Also, traffic belonging to a same flow may be queued to a same virtual core queue and processed by a same virtual core.

Claims

1. A parallel processing method for traffic having an order, the method supporting virtual core automatic scaling in a virtual environment having at least one virtual machine (VM) and at least one virtual core queue for each VM, comprising:

identifying a flow of the traffic according to attributes of the traffic;

determining a virtual core queue to which the traffic is mapped according to the identified flow and queuing the traffic to the determined virtual core queue; and

transferring the traffic from the determined virtual core queue to a VM corresponding to the determined virtual core queue through a virtual switch,

wherein traffic belonging to a same flow is queued to a same virtual core queue and processed by a same virtual core, in order to maintain the order of the traffic to be processed and not to cause a re-ordering of the traffic having the order, and

wherein the virtual core queue to which the traffic is queued is selected from virtual core queues belonging to a VM allocated to the identified flow based on at least one of a power saving policy or an operation switching policy.

2. The parallel processing method according to claim 1, wherein, in the identifying the flow of the traffic, the flow of the traffic is identified based on a hash function value for specific information in a header of a packet belonging to the traffic.

3. The parallel processing method according to claim 2, wherein the specific information in the header of the packet belonging to the traffic is a combination of at least one field value of a source IP address (SIP), a destination IP address (DIP), a source port (SPORT), a destination port (DPORT), and a protocol version (PROTOCOL).

4. The parallel processing method according to claim 1, wherein, in the queuing the traffic, the traffic is stored in a data buffer, and a pointer pointing to a location where the traffic is stored in the data buffer is queued to the determined virtual core queue.

5. The parallel processing method according to claim 1, wherein, in the queuing the traffic, the traffic is queued to the virtual queue allocated to the identified flow when the virtual core queue allocated to the identified flow exists;

and the traffic is queued to a virtual core queue selected from the virtual core queues belonging to the VM allocated to the identified flow or a newly created virtual core queue when the virtual core queue allocated to the identified flow does not exist.

6. The parallel processing method according to claim 1, wherein

according to the power saving policy, the flow of the traffic is mapped to a virtual core queue in which a largest amount of traffic is queued among existing virtual core queues having a length that does not exceed a threshold value, and

according to the operation switching policy, the flow of the traffic is mapped to a virtual core queue in which a least amount of traffic is queued among the existing virtual core queues having the length that does not exceed the threshold value.

7. The parallel processing method according to claim 1, wherein, in the transferring the traffic, the traffic is transferred to a virtual core corresponding to the determined virtual core queue in a VM to which the determined virtual core queue belongs.

8. The parallel processing method according to claim 7, wherein the traffic is transferred to the virtual core corresponding to the determined virtual core queue in the VM to which the determined virtual core queue belongs based on meta information included in the traffic.

9. A parallel processing apparatus for traffic having an order, the apparatus supporting virtual core automatic scaling, comprising at least one processor configured to execute instructions to implement:

a virtual machine (VM) manager for creating and managing at least one VM;

a virtual core queue manager for creating and managing at least one virtual core queue for each VM;

a virtual core queue mapper for identifying a flow of the traffic according to attributes of the traffic, determining a virtual core queue to which the traffic is mapped according to the identified flow, and queuing the traffic to the determined virtual core queue; and

at least one virtual switch for transferring the traffic from the determined virtual core queue to a VM corresponding to the determined virtual core queue,

wherein traffic belonging to a same flow is queued to a same virtual core queue and processed by a same virtual core, in order to maintain the order of the traffic to be processed and not to cause a re-ordering of the traffic having the order, and

wherein the virtual core queue to which the traffic is queued is selected from virtual core queues belonging to a VM allocated to the identified flow based on at least one of a power saving policy or an operation switching policy.

10. The parallel processing apparatus according to claim 9, wherein the virtual core queue mapper identifies the flow of the traffic based on a hash function value for specific information in a header of a packet belonging to the traffic.

11. The parallel processing apparatus according to claim 10, wherein the specific information in the header of the packet belonging to the traffic is a combination of at least one field value of a source IP address (SIP), a destination IP address (DIP), a source port (SPORT), a destination port (DPORT), and a protocol version (PROTOCOL).

12. The parallel processing apparatus according to claim 10, further comprising a flow table defining relationship between traffics and flows, wherein the virtual core queue mapper identifies the flow of the traffic based on the flow table.

13. The parallel processing apparatus according to claim 9, further comprising a virtual switch manager managing the at least one virtual switch, wherein the virtual switch manager adds a new virtual switch, or deactivates a part of the at least one virtual switch according to a processing load of the at least one virtual switch.

14. The parallel processing apparatus according to claim 9, wherein the virtual core queue mapper stores the traffic in a data buffer, and queues a pointer pointing to a location where the traffic is stored in the data buffer to the determined virtual core queue.

15. The parallel processing apparatus according to claim 9, wherein the virtual core queue mapper queues the traffic to the virtual queue allocated to the identified flow when the virtual core queue allocated to the identified flow exists;

and queues the traffic to a virtual core queue selected from the virtual core queues belonging to the VM allocated to the identified flow or a newly created virtual core queue when the virtual core queue allocated to the identified flow does not exist.

16. The parallel processing apparatus according to claim 15, wherein

according to the power saving policy, the flow of the traffic is mapped to a virtual core queue in which a largest amount of traffic is queued among existing virtual core queues having a length that does not exceed a threshold value, and

according to the operation switching policy, the flow of the traffic is mapped to a virtual core queue in which a least amount of traffic is queued among the existing virtual core queues having the length that does not exceed the threshold value.

17. The parallel processing apparatus according to claim 9, wherein the VM corresponding to the determined virtual core queue includes a virtual network interface card (vNIC) and a virtual distributor (vDistributor) which transfer the traffic transferred from the virtual switch to the determined virtual core queue.

18. The parallel processing apparatus according to claim 17, wherein the traffic is transferred to a virtual core corresponding to the determined virtual core queue in a VM to which the determined virtual core queue belongs based on meta information included in the traffic.