Applying policies to schedule network bandwidth among virtual machines

• US 9,471,348 B2
• Filed: 07/01/2013
• Issued: 10/18/2016
• Est. Priority Date: 05/31/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Chau, Hoyuen, Zuo, Yue, Vo, Hoi Huu, Arafeh, Samer N., Divakara, Vivek P., Deng, Yimin, Foltz, Forrest Curtis, Bhanu, Vivek
• Primary Examiner(s): Wong, Warner
• Application Number:
  US13/932,952
• Publication Number:
  US 20130298123A1
• Time in Patent Office: 1,205 Days
• Field of Search: 370/230, 370/235, 370/235.1, 370/237, 370254-255, 370/395.4, 709/201, 709/225, 709/226, 709/248
• US Class Current: 1/1
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: H04L 41/0896 Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76
• CPC Class Codes: H04L 41/0897 by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/215 using token-bucket
• CPC Class Codes: H04L 65/80 Responding to QoS

Abstract

Computerized methods, systems, and computer-storage media for allowing virtual machines (VMs) residing on a common physical node to fairly share network bandwidth are provided. Restrictions on resource consumption are implemented to ameliorate stressing the network bandwidth or adversely affecting the quality of service (QoS) guaranteed to tenants of the physical node. The restrictions involves providing a scheduler that dynamically controls networking bandwidth allocated to each of the VMs as a function of QoS policies. These QoS policies are enforced by controlling a volume of traffic being sent from the VMs. Controlling traffic includes depositing tokens into token-bucket queues assigned to the VMs, respectively. The tokens are consumed as packets pass through the token-bucket queues. Upon consumption, packets are held until sufficient tokens are reloaded to the token-bucket queues.

Claims

1. One or more computer storage memory having computer-executable instructions embodied thereon that, when executed, perform a method for metering various data flows of packets originating from a plurality of virtual machines (VMs) instantiated on a single physical node, the method comprising:

reading one or more first policies contained within a first service agreement, wherein the contract first service agreement governs allocation of network resources to a first client;

allocating a first amount of network bandwidth to a first VM, of the plurality of VMs, associated with the first client based on the one or more first policies of the first service agreement, wherein a size of the first amount of network bandwidth is based, at least in part, upon a first weighting associated with the first VM relative to, at least, a second weighting associated with a second VM of the plurality of VMs, and wherein the first weighting and the second weighting are specified by the first service agreement and a second service agreement, respectively;

detecting that the first VM is in an active state based on a depletion of the first amount of network bandwidth, wherein the first amount of network bandwidth diminishes as a function of a rate of the data flow from the first VM; and

allocating a second amount of network bandwidth to the first VM based, at least in part, on the one or more first policies and a remaining portion of the first amount of network bandwidth.

2. The one or more computer-storage memory of claim 1, the method further comprising:

reading one or more second policies contained within the second service agreement, wherein the second service agreement governs allocation of network resources to a second client;

allocating a third amount of network bandwidth to the second VM, of the plurality of VMs, associated with the second client based on the one or more second policies of the second service agreement;

detecting that the second VM is in an active state based on a depletion of the third amount of network bandwidth, wherein the third amount of network bandwidth diminishes as a function of a rate of the data flow from the second VM; and

allocating a fourth amount of network bandwidth to the second VM based, at least in part, on the one or more second policies and a remaining portion of the third amount of network bandwidth.

3. A computer system capable of metering a rate at which packets from a virtual machine are injected into a network utilizing one or more policies specified by tenants of a data center, the computer system comprising a computer storage medium having a plurality of computer software components embodied thereon, the computer software components comprising:

a first VM instantiated on a physical node, the first VM associated with a first client;

a second VM instantiated on the physical node, the second VM associated with a second client;

a first queue associated with the first VM;

a second queue associated with the second VM; and

a scheduler that reads a first service agreement and a second service agreement associated with the first client and the second client, respectively, that allocates a first amount of bandwidth to the first VM as a function of one or more first policies contained within the first service agreement, and that allocates a second amount of bandwidth to the second VM as a function of one or more second policies within the second service agreement,

wherein the first VM transmits a first data flow of packets to the first queue prior to the packets of the first data flow being released from the physical node,

wherein the second VM transmits a second data flow of packets to the second queue prior to the packets of the second data flow being released from the physical node,

wherein, upon receiving the first data flow, the first queue reduces the first amount of bandwidth by a prescribed amount for each packet within the first data flow,

wherein, upon receiving the second data flow, the second queue reduces the second amount of bandwidth by a prescribed amount for each packet within the second data flow.

4. The computer system of claim 3, further comprising:

a routing component that accesses a mapping between the first VM and the first queue in response to detecting that the first VM is attempting to transmit the first data flow to an endpoint external to the physical node, and that routes the first data flow to the first queue based on the mapping, wherein the routing component routes the first data flow from first VM to the second VM, bypassing the first queue, in response to detecting that the first VM is attempting to transmit the first data flow directly to the second VM.

5. The computer system of claim 4, the computer software components further comprising a network interface controller (NIC) queue that meters the transmission of the packets within the first data flow from the first queue and, in conjunction, meters the transmission of the packets within the second data flow from the second queue.

6. The computer system of claim 5, the computer software components further comprising a network interface component that receives the metered transmission of the packets within the first data flow and the metered transmission of the packets within the second data flow from the NIC queue and releases from the physical node a controlled volume of traffic that includes the packets within the first data flow and the packets within the second data flow.

7. The computer system of claim 3, wherein the first queue includes a queue component that internally enqueues a subset of packets contained within the first data flow that are received thereby in response to detecting that the first amount of bandwidth is insufficient for consumption.

8. A computerized method for imposing a limit on a rate of data flowing from a subject virtual machine (VM), the method comprising:

accepting a current allocation of bandwidth from a scheduler, wherein an amount of bandwidth provided within the current allocation is based at least in part on a service agreement associated with the subject VM and whether the subject VM is in an active state or an idle state, and

wherein determining whether the subject VM is in an active state is based on a depletion of a previous allocation of bandwidth and determining whether the subject VM is in an idle state is based on a lack of usage of the previous allocation of bandwidth,

dynamically metering a data-flow rate of the subject VM, wherein metering includes;

(a) forwarding, to a network interface component, packets transmitted by the subject VM that are received prior to the current allocation of bandwidth being substantially depleted;

(b) enqueueing packets transmitted by the subject VM that are received subsequent to the current allocation of bandwidth being substantially depleted; and

(c) forwarding the enqueued packets to the network interface component upon accepting another allocation of bandwidth from the scheduler.

9. The computerized method of claim 8, wherein the amount of bandwidth provided within the current allocation is an absolute value specified by the service agreement.

10. The computerized method of claim 8, wherein the amount of bandwidth provided within the current allocation is based further upon a first weighting associated with the subject VM relative to, at least, a second weighting associated with another VM, wherein the another VM is operating on a same physical node as the subject VM.

11. The computerized method of claim 10, wherein the first weighting is specified by the service agreement associated with the subject VM and the second weighting is specified by another service agreement associated with the another VM.

12. The computerized method of claim 8, wherein the amount of bandwidth provided within the current allocation is based further upon an amount of network bandwidth currently available to a physical node on which the subject VM is operating.

13. The computerized method of claim 8, wherein the current allocation is consumed as a function of the data flow rate rate of the subject VM.

14. The computerized method of claim 8, wherein dynamically metering the data-flow rate of the subject VM further includes:

forwarding, to the network interface component, packets transmitted by the subject VM that are being directly transmitted to another VM operating on a same physical node as the subject VM, regardless of whether the current allocation of bandwidth is substantially depleted.