Network load balancing with traffic routing

• US 7,567,504 B2
• Filed: 09/08/2003
• Issued: 07/28/2009
• Est. Priority Date: 06/30/2003
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Darling, Christopher L., Joy, Joseph M., Welland, Robert V., House, Sean B., Hydrie, Aamer
• Primary Examiner(s): HSU, ALPUS
• Application Number:
  US10/657,412
• Publication Number:
  US 20040264481A1
• Time in Patent Office: 2,150 Days
• Field of Search: 370/216, 370/231, 370/235, 370/428, 709/203, 709/227, 709/238
• US Class Current: 370/216
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1017 based on a round robin mechanism
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54
• CPC Class Codes: H04L 67/568 Storing data temporarily at an intermediate stage, e.g. caching
• CPC Class Codes: H04L 67/61 taking into account QoS or priority requirements
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

In an exemplary method implementation, a method includes: receiving a packet requesting a new connection at a forwarding component; sending the packet from the forwarding component to a classifying component; selecting, by the classifying component, a route for the new connection; and plumbing, by the classifying component, the route for the new connection by causing a new entry to be added in a local routing table of the forwarding component. In an exemplary media implementation, one or more processor-accessible media include processor-executable instructions that, when executed, enable a system to perform actions including: receiving a first packet for a connection at first forwarding functionality; plumbing a route for the connection at the first forwarding functionality; receiving a second packet for the connection at second forwarding functionality; and plumbing the route for the connection at the second forwarding functionality using a distributed session tracking table.

Claims

1. A system to load balance packetized data connections among a plurality of hosts comprising:

a forwarding component that forwards packets according to entries in a routing table;

a classifying component capable of classifying packets for the forwarding component and updating the forwarding component routing table according to packet content and host status information including a host'"'"'s health and load information and wherein the classifying component is separate from the forwarding component to enhance scalability;

a session tracking component that tracks sessions for at least one of the forwarding component and the classifying component;

a health and load handling component that is capable of receiving health and load information from a host and providing the health and load information to the classifying component; and

a high availability mechanism that provides detection of, handling of, and recovery from a failure of one or more of the forwarding component, the classifying component, the session tracking component, and the health and load handling component, and wherein once a packet in a connection has been classified subsequent packets in that connection bypass the classifying component and are sent directly to the host.

2. The system as recited in claim 1, wherein the system further comprises:

a request routing component that is capable of routing logical requests;

wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the request routing component.

3. The system as recited in claim 1, wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the forwarding component;

the high availability mechanism including capabilities for detection of a failure at the forwarding component by at least one load-balancing-aware switch, redirection of packets to at least one other forwarding component, and rebuilding of lost routes with a distributed session tracking manager.

4. The system as recited in claim 1, wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the classifying component;

the high availability mechanism including capabilities for detection of a failure at the classifying component by at least one forwarding component, redirection of packets to at least one other classifying component, and rebuilding of lost session information with a distributed session tracking manager.

5. The system as recited in claim 1, wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the session tracking component;

the high availability mechanism including capabilities for detection of a failure at the session tracking component by at least one forwarding component and/or classifying component and for distributed and redundant storage of session information.

6. The system as recited in claim 1, wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the health and load handling component;

the high availability mechanism including capabilities for detection of a failure at the health and load handling component by at least one classifying component and for rebuilding of a cache of health and load information using a message protocol.

7. The system as recited in claim 1, wherein the high availability mechanism provides detection of, handling of, and recovery from a failure of the health and load handling component;

the high availability mechanism including capabilities for redundant storing of health and load information and for authoritative storing of health and load information at hosts to which the health and load information pertains.

8. The system as recited in claim 1, wherein the forwarding component, the classifying component, the session tracking component, and the health and load handling component are resident at and executing on at least two different devices.

9. A system for highly available network load balancing infrastructure, the system comprising:

a plurality of different means for load balancing network traffic wherein once a packet is classified by a classifier means subsequent packets in that connection are forwarded by a forwarder means without additional classification wherein the classifier means is separate from the forwarder means to enhance scalability, wherein the forwarder means has a relatively greater ability to accommodate a high packet flux and the classifier means has a relatively greater ability to accommodate processing-intensive tasks;

detection means for detecting a failure of one or more of the plurality of different means for load balancing;

handling means for handling the failure; and

recovery means for recovering from the failure.

10. The system as recited in claim 9, wherein the plurality of different means for load balancing includes at least one request router means for routing packets on a request-level.

11. The system as recited in claim 9, wherein the plurality of different means for load balancing includes at least one session tracker means for tracking sessions.

12. The system as recited in claim 9, wherein the plurality of different means for load balancing includes at least one health and load handler means for handling health and load information.

13. The system as recited in claim 9, wherein the system comprises one or more processor-accessible media.

14. A network load balancing system comprising:

a first device that includes forwarding functionality; and

a second device that includes classifying functionality, the classifying functionality performing classifying for the forwarding functionality wherein once a packet has been classified subsequent packets in that connection are forwarded without further classification, and further wherein hardware of the first device differs and is separate from hardware of the second device to enhance scalability;

wherein the hardware of the first device has a relatively greater ability to accommodate a high packet flux, and the hardware of the second device has a relatively greater ability to accommodate processing-intensive tasks.

15. The network load balancing system as recited in claim 14, wherein the hardware of the first device is especially tuned for the forwarding functionality, and the separate hardware of the second device is especially tuned for the classifying functionality.

16. The network load balancing system as recited in claim 14, wherein the hardware of the first device comprises a router or switch, and the hardware of the second device comprises a personal computer or server.