System for Distributing Load Over Multiple Servers at an Internet Site

• US 20070088822A1
• Filed: 12/04/2006
• Published: 04/19/2007
• Est. Priority Date: 11/03/1995
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Coile, Brantley, Kersey, Edward, Tenereillo, Peter, Howes, Richard
• Granted Patent: US 7,730,190 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H01R 2201/04 for network, e.g. LAN connectors
• CPC Class Codes: H01R 31/005 Intermediate parts for distributing signals
• CPC Class Codes: H04L 2101/663 Transport layer addresses, e.g. aspects of transmission control protocol [TCP] or user datagram protocol [UDP] ports
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 61/00 Network arrangements, protocols or services for addressing or naming
• CPC Class Codes: H04L 61/2514 between local and global IP addresses
• CPC Class Codes: H04L 61/2564 for a higher-layer protocol, e.g. for session initiation protocol [SIP]
• CPC Class Codes: H04L 63/0254 Stateful filtering
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1017 based on a round robin mechanism
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

Disclosed is a system and method for distributing connections among a plurality of servers at an Internet site. All connections are made to a single IP address and a local director selects the server from among the plurality of servers which is to receive the connection. Thus, the DNS server is not relied upon to distribute connections, and the connection distribution scheme is not avoided when DNS is bypassed. In one embodiment, a session distribution scheme is implemented such that connections are distributed to the server in the group of servers which has the fewest connections of the group. In other embodiments, other session distribution schemes which route connections based on the predicted response times of the servers or according to a round robin scheme are used.

Claims

1. (canceled)

2. (canceled)

3. (canceled)

4. (canceled)

5. (canceled)

6. (canceled)

7. (canceled)

8. (canceled)

9. (canceled)

10. (canceled)

11. (canceled)

12. (canceled)

13. (canceled)

14. (canceled)

15. (canceled)

16. (canceled)

17. (canceled)

18. (canceled)

19. (canceled)

20. (canceled)

21. (canceled)

22. A method for distributing packets among a plurality of physical machines, comprising:

establishing a virtual machine associated with one or more physical machines;

receiving a packet associated with a session and indicating the virtual machine;

determining expected response times for each physical machine;

identifying a particular physical machine that is expected to process the packet in the shortest amount of time;

routing the packet to the particular physical machine.

23. The method of claim 22, further comprising:

determining a number of active connections for each physical machine associated with the virtual machine;

identifying a physical machine having the fewest active connections as the particular physical machine.

24. The method of claim 22, further comprising:

determining a time delay interval for each physical machine associated with the virtual machine, the time delay interval being a time between receipt of a packet and generation of a response to the packet;

identifying a physical machine having the shortest time delay interval as the particular physical machine.

25. The method of claim 24, further comprising:

identifying a physical machine not having the shortest time delay interval as the particular physical machine;

establishing a new time delay interval for the particular physical machine.

26. The method of claim 22, further comprising:

determining whether a physical machine has been newly activated;

successively selecting the physical machines as the particular physical machine in response to a newly activated physical machine until the newly activated physical machine has stabilized.

27. The method of claim 22, further comprising:

determining whether a physical machine has failed;

overriding identification of the failed physical machine as the particular physical machine.

28. The method of claim 22, further comprising:

determining whether the packet associated with the session has an originating address associated with a previous session;

identifying a same physical machine as identified for the previous session as the particular physical machine.

29. The method of claim 28, wherein the same physical machine is identified if the session falls within a set period of time from the previous session.

30. A system for distributing packets among a plurality of physical machines, comprising:

means for establishing a virtual machine associated with one or more physical machines;

means for receiving a packet associated with a session and indicating the virtual machine;

means for determining expected response times for each physical machine;

means for identifying a particular physical machine that is expected to process the packet in the shortest amount of time;

means for routing the packet to the particular physical machine.

31. The system of claim 30, further comprising:

means for determining a number of active connections for each physical machine associated with the virtual machine;

means for identifying a physical machine having the fewest active connections as the particular physical machine.

32. The system of claim 30, further comprising:

means for determining a time delay interval for each physical machine associated with the virtual machine, the time delay interval being a time between receipt of a packet and generation of a response to the packet;

means for identifying a physical machine having the shortest time delay interval as the particular physical machine.

33. The system of claim 32, further comprising:

means for identifying a physical machine not having the shortest time delay interval as the particular physical machine;

means for establishing a new time delay interval for the particular physical machine.

34. The system of claim 30, further comprising:

means for determining whether a physical machine has been newly activated;

means for successively selecting the physical machines as the particular physical machine in response to a newly activated physical machine until the newly activated physical machine has stabilized.

35. The system of claim 30, further comprising:

means for determining whether a physical machine has failed;

means for overriding identification of the failed physical machine as the particular physical machine.

36. The system of claim 30, further comprising:

means for determining whether the packet associated with the session has an originating address associated with a previous session;

means for identifying a same physical machine as identified for the previous session as the particular physical machine.

37. The system of claim 36, wherein the same physical machine is identified if the session falls within a set period of time from the previous session.

38. A computer readable medium including code for distributing packets among a plurality of physical machines, the code operable to:

establish a virtual machine associated with one or more physical machines;

receive a packet associated with a session and indicating the virtual machine;

determine expected response times for each physical machine;

identify a particular physical machine that is expected to process the packet in the shortest amount of time;

route the packet to the particular physical machine.

39. The computer readable medium of claim 38, wherein the code is further operable to:

determine a number of active connections for each physical machine associated with the virtual machine;

identify a physical machine having the fewest active connections as the particular physical machine.

40. The computer readable medium of claim 38, wherein the code is further operable to:

determine a time delay interval for each physical machine associated with the virtual machine, the time delay interval being a time between receipt of a packet and generation of a response to the packet;

identify a physical machine having the shortest time delay interval as the particular physical machine.

41. The computer readable medium of claim 40, wherein the code is further operable to:

identify a physical machine not having the shortest time delay interval as the particular physical machine;

establish a new time delay interval for the particular physical machine.

42. The computer readable medium of claim 38, wherein the code is further operable to:

determine whether a physical machine has been newly activated;

successively select the physical machines as the particular physical machine in response to a newly activated physical machine until the newly activated physical machine has stabilized.

43. The computer readable medium of claim 38, wherein the code is further operable to:

determine whether a physical machine has failed;

override identification of the failed physical machine as the particular physical machine.

44. The computer readable medium of claim 38, wherein the code is further operable to:

determine whether the packet associated with the session has an originating address associated with a previous session;

identify a same physical machine as identified for the previous session as the particular physical machine.

45. The computer readable medium of claim 44, wherein the same physical machine is identified if the session falls within a set period of time from the previous session.

46. A system for distributing packets among a plurality of physical machines, comprising:

a plurality of physical machines;

a local director operable to establish a virtual machine associated with the plurality of physical machines, the local director operable to receive a packet associated with a session and indicating the virtual machine, the local director operable to determine expected response times for each physical machine, the local director operable to identify a particular physical machine that is expected to process the packet in the shortest amount of time, the local director operable to route the packet to the particular physical machine identified in the session distribution scheme.

47. The system claim 46, wherein the local director is operable to determine a number of active connections for each physical machine associated with the virtual machine, the local director operable to identify a physical machine having the fewest active connections as the particular physical machine.

48. The system of claim 46, wherein the local director is operable to determine a time delay interval for each physical machine associated with the virtual machine, the time delay interval being a time between receipt of a packet and generation of a response to the packet, the local director operable to identify a physical machine having the shortest time delay interval as the particular physical machine.

49. The system of claim 48, wherein the local director is operable to periodically age and update the time delay interval for each physical machine.

50. The system of claim 48, wherein the local director is operable to identify a physical machine not having the shortest time delay interval as the particular physical machine, the local director operable to establish a new time delay interval for the particular physical machine.

51. The system of claim 46, wherein the local director is operable to determine whether a physical machine has been newly activated, the local director operable to successively select the physical machines as the particular physical machine in response to a newly activated physical machine until the newly activated physical machine has stabilized.