Distributed Traffic Controller for Network Data

• US 20080104273A1
• Filed: 10/26/2007
• Published: 05/01/2008
• Est. Priority Date: 11/10/1999
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Bruck, Jehoshua, LeMahieu, Paul, Fan, Chenggong, Bohosslan, Vasken, Love, Philip
• Granted Patent: US 7,613,827 B2
• Time in Patent Office: Days
• US Class Current: 709/238
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 61/5007 Internet protocol [IP] addresses
• 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 67/101 based on network conditions
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer

Abstract

A distributed gateway for controlling computer network data traffic dynamically reconfigures traffic assignments among multiple gateway machines for increased network availability. If one of the distributed gateway machines becomes unavailable, traffic assignments are moved among the multiple machines such that network availability is substantially unchanged. The machines of the distributed gateway form a cluster and communicate with each other using a Group Membership protocol word such that automatic, dynamic traffic assignment reconfiguration occurs in response to machines being added and deleted from the cluster, with no loss in functionality for the gateway overall, in a process that is transparent to network users, thereby providing a distributed gateway functionality that is scalable. Operation of the distributed gateway remains consistent as machines are added and deleted from the cluster. A scalable, distributed, highly available, load balancing network gateway is thereby provided, having multiple machines that function as a front server layer between the network and a back-end server layer having multiple machines functioning as Web file servers, FTP servers, or other application servers. The front layer machines comprise a server cluster that performs fail-over and dynamic load balancing for both server layers.

Claims

1-8. -8. (canceled)

9. A method of operating a server computer of a plurality of server computers that comprise a front layer of server computers for controlling data traffic of a computer network, said plurality of server computers being members of a first subnet of network addresses over which network data traffic is sent and received, the method comprising:

receiving network data traffic through a network interface that permits communication between the server computer and other server computers of said plurality of server computers;

communicating with the plurality of server computers state sharing information with a dynamic reconfiguration protocol that permits reassignment of any of a plurality of virtual network addresses to any of a plurality of primary network addresses among the front layer server computers and includes state information sharing and load information sharing among the front layer servers; and

communicating with a plurality of network computers that are members of a second subnet of network addresses to send and receive said network data traffic.

10. A method as defined in claim 9, wherein communicating with the plurality of server computers comprises sending data using a Reliable Message layer scheme that comprises a token data packet and one or more data carriage packets, wherein the token data packet specifies the number of data carriage packets that together comprise a Reliable Message packet and wherein the data carriage packets include data relating to state information and data traffic load information about each of the front layer server computers.

11. A method as defined in claim 9, further including:

configuring an operating system of the server computer such that all network addresses in a pool of addresses assigned to the plurality of server computers of the first subnet are assigned to the server computer;

generating a gratuitous address resolution protocol (ARP) message in response to an address reassignment of the server computer and communicating the ARP message to the other server computers of the first subnet; and

blocking the sending of an ARP acknowledgment message to the other server computers of the first subnet for any received gratuitous ARP message, thereby inhibiting reboot operation of the respective server computers of said first subnet and ensuring that each server computer of said first subnet is unaware of any duplicate assignment of network address numbers.

12. A method as defined in claim 9, further including operating said server computer as an authoritative node of the first subnet to ensure symmetric routing of network data traffic to and from the first subnet.

13. A method as defined in claim 12, wherein said operating said server computer to ensure symmetric traffic routing comprises:

receiving a data request from a responding server computer of the first subnet, wherein the data request was initially received at the responding server computer;

determining the authoritative node for responding to the data request;

identifying a server computer in the first subnet that will handle the data traffic associated with the data request and forwarding the data request to the identified server computer for handling;

receiving a reply message from a server computer of the first subnet that is operating as a default reply node to a network computer of said second subnet that is responding to the data request; and

forwarding the reply message to a server computer of the first subnet that will ensure symmetric routing of the data request and reply message with respect to the plurality of server computers of the first subnet.

14. A method as defined in claim 13, further comprising forwarding assignment information to the server computer of the first subnet that was operating as the default reply node for the data request, wherein the assignment information includes information that the default reply node can use to directly forward response messages from the network computer of the second subnet to the server computer of the first subnet that will ensure symmetric routing.

15. A program product for use in a computer that executes program steps recorded in a computer-readable media to perform a method of operating the computer, the program product comprising:

a recordable media;

a plurality of computer-readable instructions executable by the computer to perform a method of operating a server computer of a plurality of server computers that comprise a front layer of server computers for controlling data traffic of a computer network, said plurality of server computers being members of a first subnet of network addresses over which network data traffic is sent and received, the method comprising;

receiving network data traffic through a network interface that permits communication between the server computer and other server computers of said plurality of server computers;

communicating with the plurality of server computers state sharing information with a dynamic reconfiguration protocol that permits reassignment of any of a plurality of virtual network addresses to any of a plurality of primary network addresses among the front layer server computers and includes state information sharing and load information sharing among the front layer servers; and

communicating with a plurality of network computers that are members of a second subnet of network addresses to send and receive said network data traffic.

16. A program product as defined in claim 15, wherein communicating with the plurality of server computers comprises sending data using a Reliable Message layer scheme that comprises a token data packet and one or more data carriage packets, wherein the token data packet specifies the number of data carriage packets that together comprise a Reliable Message packet and wherein the data carriage packets include data relating to state information and data traffic load information about each of the front layer server computers.

17. A program product as defined in claim 15, further including:

configuring an operating system of the server computer such that all network addresses in a pool of addresses assigned to the plurality of server computers of the first subnet are assigned to the server computer;

generating a gratuitous address resolution protocol (ARP) message in response to an address reassignment of the server computer and communicating the ARP message to the other server computers of the first subnet; and

blocking the sending of an ARP acknowledgment message to the other server computers of the first subnet for any received gratuitous ARP message, thereby inhibiting reboot operation of the respective server computers of the first subnet and ensuring that each server computer of the first subnet is unaware of any duplicate assignment of network address numbers.

18. A program product as defined in claim 15, further including operating said server computer as an authoritative node of the first subnet to ensure symmetric routing of network data traffic to and from the first subnet.

19. A program product as defined in claim 18, wherein operating to ensure symmetric traffic routing comprises:

receiving a data request from a responding server computer of the first subnet, wherein the data request was initially received at the responding server computer;

determining the authoritative node for responding to the data request;

identifying a server computer in the first subnet that will handle the data traffic associated with the data request and forwarding the data request to the identified server computer for handling;

receiving a reply message from a server computer of the first subnet that is operating as a default reply node to a network computer of said second subnet that is responding to the data request; and

forwarding the reply message to a server computer of the first subnet that will ensure symmetric routing of the data request and reply message with respect to the plurality of server computers of the first subnet.

20. A program product as defined in claim 19, further comprising forwarding assignment information to the server computer of the first subnet that was operating as the default reply node for the data request, wherein the assignment information includes information that the default reply node can use to directly forward response messages from the network computer of the second subnet to the server computer of the first subnet that will ensure symmetric routing.