Software defined network-based load balancing for physical and virtual networks

• US 9,407,560 B2
• Filed: 03/15/2013
• Issued: 08/02/2016
• Est. Priority Date: 03/15/2013
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Crisan, Daniel, DeCusatis, Casimer, Gusat, Mitch, Minkenberg, Cyriel
• Primary Examiner(s): Oh, Andrew
• Application Number:
  US13/834,117
• Publication Number:
  US 20140269288A1
• Time in Patent Office: 1,236 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: H04L 41/5019 Ensuring fulfilment of SLA
• CPC Class Codes: H04L 41/5022 by giving priorities, e.g. assigning classes of service
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 47/11 Identifying congestion
• CPC Class Codes: H04L 47/115 using a dedicated packet
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/122 by diverting traffic away from congested entities
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04L 47/2425 for supporting services specification, e.g. SLA

Abstract

A method of monitoring transmission of data in a network is provided. The method includes: receiving data packets for one or more data flows by at least one of a plurality of network switches; sampling queue data in each of the plurality of network switches, the sampled queue data taken from at least one queue attached to a switch port to which the one or more data packets have been forwarded; and transmitting the sampled queue data to a logically centralized network controller connected to the plurality of switches, the network controller configured to control the plurality of network switches, the network controller configured to determine whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more switches based on the sampled queue data and send a control message to one or more data flow sources based on the congestion, unfairness or SLA infringement condition.

Claims

1. A method of monitoring transmission of data in a network, comprising:

receiving data packets for one or more data flows by at least one of a plurality of network switches;

sampling queue data in each of the plurality of network switches receiving the one or more data packets, the sampled queue data taken from at least one queue attached to a switch port to which the one or more data packets have been forwarded, wherein the sampling of queue data from the switch port is performed by the network switch that has the switch port;

encapsulating, by the network switch that performs the sampling, the sampled queue data in a control packet configured for communication between the plurality of network switches and a logically centralized network controller, the control packet configured to be used by the network controller to control switch configuration and routing; and

transmitting the control packet including the sampled queue data by each network switch to the network controller connected to the plurality of switches, the network controller configured to control the plurality of network switches, the network controller configured to determine whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more switches based on the sampled queue data, the network controller configured to select a time period during which the congestion, unfairness or SLA infringement condition was detected, aggregate all of the sampled queue data that was sampled during the time period, and send a control message to all data flow sources responsible for the congestion, unfairness or SLA infringement condition, the control message setting a globally fair transmission rate for all data flows for which the congestion, unfairness or SLA condition was detected, the control message sent to each of the data flow sources so that all of the data flow sources responsible for the congestion, unfairness or SLA infringement condition receive the control message within a time interval that is smaller than the time period.

2. The method of claim 1, wherein the globally fair transmission rate is calculated based on a number of data flows affected by the congestion, unfairness or SLA condition, and based on a severity of the congestion, unfairness or SLA condition.

3. The method of claim 1, wherein the network controller is configured to determine whether the congestion, unfairness or SLA infringement condition exists based on a rate of change of depth of the at least one queue, the depth corresponding to an amount of the at least one queue that is filled with data.

4. The method of claim 1, further comprising receiving the control message from the network controller by the at least one of a plurality of network switches, wherein the control message is a global control message sent to each of the plurality of network switches based on aggregated sampled queue data from the plurality of network switches for a selected time period, the global control message including rate control calculated based on the aggregated sampled queue data.

5. The method of claim 1, wherein transmitting the sampled queue data includes sending the sample data packet to the network controller over a secure link, the secure link established by the network controller to configure the switch and establish communication properties of links between network switches.

6. A method of monitoring transmission of data in a network, comprising:

receiving, by a centralized network controller, a control packet encapsulating sampled queue data from each of a plurality of network switches, the network controller configured to control the plurality of network switches, the sampled queue data taken from at least one queue attached to a switch port to which one or more data packets of a data flow have been forwarded, the sampling of queue data from the switch port performed by the network switch that has the switch port, the control packet configured to be used by the network controller to control switch configuration and routing;

determining, by the network controller, whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more network switches for one or more data flows based on the sampled queue data; and

based on determining that the congestion, unfairness or SLA infringement condition exists for the one or more data flows, selecting a time period during which the congestion, unfairness or SLA infringement condition was detected, aggregating all of the sampled queue data that was sampled during the time period and sending a control message via a control packet to all data flow sources responsible for the congestion, unfairness or SLA infringement condition to reduce an amount of congestion, unfairness or SLA non-compliance in the network switch, the control message setting a globally fair transmission rate for all data flows for which the congestion, unfairness or SLA condition was detected, the control message sent to each of the data flow sources so that all of the data flow sources responsible for the congestion, unfairness or SLA infringement condition receive the control message within a time interval that is smaller than the time period.

7. The method of claim 6, wherein the plurality of network switches is a plurality of OpenFlow switches and the network controller is an OpenFlow network controller.

8. The method of claim 6, wherein the plurality of network switches are configured for quantized congestion control (QCN) monitoring, which is configured to generate a congestion notification message (CNM).

9. The method of claim 6, wherein the control message is sent via a multicast message to each of the plurality of data flow sources within the time interval.

10. The method of claim 6, wherein determining includes aggregating sampled queue data for each data flow over a congestion epoch, and determining whether the congestion, unfairness or SLA infringement condition exists based on a number of congested data flows and at least one of congestion severity, an unfairness severity and a SLA infringement severity.

11. The method of claim 10, wherein the control message includes a transmission rate set for each congested data flow.

12. The method of claim 6, wherein sending the control message includes sending the control message to a plurality of data flow sources via a multicast function.

13. The method of claim 6, wherein the control message includes a transmission rate that is set individually for each of the one or more data flow sources.

14. An apparatus for controlling a plurality of switches in a network, comprising:

a memory having computer readable computer instructions; and

a processor for executing the computer readable instructions, the instructions including;

receiving a control packet encapsulating sampled queue data from each of a plurality of network switches, the sampled queue data taken from at least one queue attached to a switch port to which one or more data packets of a data flow have been forwarded, the sampling of queue data from the switch port performed by each of the plurality of network switches that has the switch port, the control packet configured to be used to control switch configuration and routing;

determining whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more network switches for one or more data flows based on the sampled queue data; and

based on determining that the congestion, unfairness or SLA infringement condition exists for the one or more data flows, selecting a time period during which the congestion, unfairness or SLA infringement condition was detected, aggregating all of the sampled queue data that was sampled during the time period, and sending a control message via a control packet to all data flow sources responsible for the congestion, unfairness or SLA infringement condition to reduce an amount of congestion, unfairness or SLA non-compliance in the network, the control message setting a globally fair transmission rate for all data flows for which the congestion, unfairness or SLA condition was detected, the control message sent to each of the data flow sources so that all of the data flow sources responsible for the congestion, unfairness or SLA infringement condition receive the control message within a time interval that is smaller than the time period.

15. The apparatus of claim 14, wherein the network is a Software Defined Network (SDN).

16. The apparatus of claim 14, wherein determining includes aggregating sampled queue data for each data flow over a congestion epoch, and determining whether the congestion, unfairness or SLA infringement condition exists based on a number of congested data flows and at least one of congestion severity, an unfairness severity and a SLA infringement severity.

17. The apparatus of claim 16, wherein the control message includes a maximal (upper bound) transmission rate set for each congested data flow.

18. The apparatus of claim 14, wherein sending the control message includes sending the control message to each source via one of a multicast and a unicast.

19. The apparatus of claim 14, wherein the control message is sent to a plurality of data flow sources within a time period that is less than a time period around which the congestion, unfairness or SLA infringement condition was detected.

20. A computer program product for monitoring transmission of data in a network, the computer program product comprising:

a non-transitory tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising;

receiving a control packet encapsulating sampled queue data from each of a plurality of network switches at a centralized network controller configured to control the plurality of network switches, the sampled queue data taken from at least one queue attached to a switch port to which one or more data packets of a data flow have been forwarded, the sampling of queue data from the switch port performed by each of the plurality of network switches that has the switch port, the control packet configured to be used by the network controller to control switch configuration and routing;

determining whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more network switches for one or more data flows based on the sampled queue data; and

based on determining that a congestion, unfairness or SLA infringement condition exists for the one or more data flows, selecting a time period during which the congestion, unfairness or SLA infringement condition was detected, aggregating all of the sampled queue data that was sampled during the time period, and sending a control message via a control packet to all data flow sources responsible for the congestion, unfairness or SLA infringement condition to reduce an amount of congestion, unfairness or SLA non-compliance in the one or more network switches, the control message setting a globally fair transmission rate for all data flows for which the congestion, unfairness or SLA condition was detected, the control message sent to each of the data flow sources so that all of the data flow sources responsible for the congestion, unfairness or SLA infringement condition receive the control message within a time interval that is smaller than the time period.

21. The computer program product of claim 20, wherein determining includes aggregating sampled queue data for each data flow over a congestion epoch, and determining whether the congestion, unfairness or SLA infringement condition exists based on a number of congested data flows and at least one of congestion severity, an unfairness severity and a SLA infringement severity.

22. The computer program product of claim 21, wherein the control message includes a maximal (upper bound) transmission rate set for each congested data flow.

23. The computer program product of claim 20, wherein the control message is sent to a plurality of data flow sources within a time period that is less than a time period around which the congestion, unfairness or SLA infringement condition was detected.

24. A computer program product for monitoring transmission of data in a network, the computer program product comprising:

a non-transitory tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising;

receiving data packets for one or more data flows by at least one of a plurality of network switches;

sampling queue data in each of the plurality of network switches receiving the one or more data flows, the sampled queue data taken from at least one queue attached to a switch port to which the one or more data packets have been forwarded, wherein the sampling of queue data from the switch port is performed by each of the plurality of network switches that has the switch port;

encapsulating, by the network switch that performs the sampling, the sampled queue data in a control packet configured for communication between the plurality of network switches and a logically centralized network controller, the control packet configured to be used by the network controller to control switch configuration and routing; and

transmitting the control packet including the sampled queue data by each network switch to the network controller connected to the plurality of switches, the network controller configured to control the plurality of network switches, the network controller configured to determine whether a congestion, unfairness or Service Level Agreement (SLA) infringement condition exists in one or more switches based on the sampled queue data, the network controller configured to select a time period during which the congestion, unfairness or SLA infringement condition was detected, aggregate all of the sampled queue data that was sampled during the time period, and send a control message to all data flow sources responsible for the congestion, unfairness or SLA infringement condition, the control message setting a globally fair transmission rate for all data flows for which the congestion, unfairness or SLA condition was detected, the control message sent to each of the data flow sources so that all of the data flow sources responsible for the congestion, unfairness or SLA infringement condition receive the control message within a time interval that is smaller than the time period.

25. The method of claim 24, wherein the plurality of network switches is a plurality of OpenFlow switches and the network controller is an OpenFlow network controller.