Application based active-active data center network using route health injection and IGP

• US 7,769,886 B2
• Filed: 02/25/2005
• Issued: 08/03/2010
• Est. Priority Date: 02/25/2005
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Naseh, Zeeshan, Gundi, Vinay
• Primary Examiner(s): Vaughn, Jr.; William C.
• Assistant Examiner(s): BENGZON, GREG C
• Application Number:
  US11/066,955
• Publication Number:
  US 20060195607A1
• Time in Patent Office: 1,985 Days
• Field of Search: 709238-241, 709/225, 370229-238, 370/351, 370396-397, 370217-219
• 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 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1006 with static server selection, e.g. the same server being selected for a specific client
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30

Abstract

A distributed data center topology having at least a pair of active data centers that can recover from a disaster at one of the data centers and achieves load balancing using IGP (Interior Gateway Protocol) between data centers. The distributed data centers use virtual IP addresses, route health injection and Border Gateway Protocol (BGP) for business continuance, disaster recovery and load balancing. Active applications are deployed at each data center to provide a logical active/standby configuration for certain applications. Alternatively, active applications are deployed at both sites and BGP routes traffic to the closest data center edge router. Load balancing occurs over an internal IGP link between sites.

Claims

1. A data center topology coupled to the internet, the data center topology comprising:

first and second active data centers, the first and second data centers being coupled to the Internet via respective first and second edge routers, wherein each of the first and second data centers comprise computer resources, wherein the first and second edge routers are coupled directly to a first Internet service provider (ISP) using a Border Gateway Protocol (BGP), wherein the second edge router is coupled directly to a second ISP using the BGP, wherein the first edge router is not coupled directly to the second ISP, wherein the data centers are in an autonomous system having a common administration and common routing policies, each edge router broadcasting a common Internet Protocol (IP) address, the edge router linking the first and second data centers to each other using an Interior Gateway Protocol (IGP);

each of said edge routers having a multilayer switch feature card (MSFC);

each edge router configured to redistribute route health injection (RHI) routes into open shortest path first (OSPF) and to summarize a host route on said MSFC;

each edge router further adapted to distribute OSPF into their BGP process using a prefix-list and to update neighboring routers with said routes;

means for implementing application-based load balancing between said first and second data centers; and

means for detecting a disaster at one of said first and second data centers by detecting a route being removed from a routing table.

2. The data center topology of claim 1 wherein each of the first and second data centers is logically partitioned into an active and a standby site.

3. The data center topology of claim 1 wherein each of the first and second edge routers are coupled by an internal network to a router and content switching module, said content switching module and said router sharing a common chassis for supporting virtual IP addresses.

4. The data center topology of claim 3 wherein a same virtual IP address (VIP) to represent the server cluster at each data center for each application.

5. The data center topology of claim 1 further comprising means for translating private virtual IP addresses to a public IP address.

6. A method for load balancing and recovering from a disaster comprising:

hosting active applications and serving clients from first and second active data centers in an autonomous system having a common administration and common routing policies, the first and second data centers being coupled to the Internet via respective first and second edge routers, wherein each of the first and second data centers comprise computer resources, wherein the first and second edge routers are coupled directly to a first Internet service provider (ISP) using a Border Gateway Protocol (BGP), wherein the second edge router is coupled directly to a second ISP using the BGP, wherein the first edge router is not coupled directly to the second ISP, wherein the data centers are in an autonomous system having a common administration and common routing policies, each edge router broadcasting a common Internet Protocol (IP) address, the edge router linking the first and second data centers to each other using an Interior Gateway Protocol (IGP);

each of said edge routers having a multilayer switch feature card (MSFC);

each edge router configured to redistribute route health injection (RHI) routes into open shortest path first (OSPF) and to summarize a host route on said MSFC;

each edge router further adapted to distribute OSPF into their BGP process using a prefix-list and to update neighboring routers with said routes;

configuring route summarization in BGP;

configuring BGP route redistribution and route preference to update neighboring routers; and

using application-based load balancing on a direct link between said first and second data centers.

7. The method of claim 6 further comprising using Exterior Border Gateway Protocol (E-BGP) for interautonomous system routing and to exchange network routing information with other networks.

8. The method of claim 7 further comprising the step of controlling incoming and outgoing routes in BGP by the using weights.

9. The method of claim 7 further comprising probing the health of the servers in each of the first and second data centers and application daemons that run on the servers.

10. A data center system coupled to the internet, the data center comprising:

first and second active data centers, the first and second data centers being coupled to the Internet via respective first and second edge routers, wherein each of the first and second data centers comprise computer resources, wherein the first and second edge routers are coupled directly to a first Internet service provider (ISP) using a Border Gateway Protocol (BGP), wherein the second edge router is coupled directly to a second ISP using the BGP, wherein the first edge router is not coupled directly to the second ISP, wherein the data centers are in an autonomous system having a common administration and common routing policies, each edge router broadcasting a common Internet Protocol (IP) address, each of said edge routers having a multilayer switch feature card (MSFC), and each of the first and second edge routers configured to;

redistribute route health injection (RHI) routes into open shortest path first (OSPF);

summarize a host route on said MSFC;

distribute routes from open shortest path first (OSPF) into the BGP process of each edge router using a prefix-list to update neighboring routers with said routes; and

utilize an Interior Gateway Protocol (IGP) link for coupling said first and second data centers.

11. The data center system of claim 10 wherein one of said the first and second data centers may change its active status by an IGP cost modification.

12. The data center system of claim 10 wherein each of the first and second data centers comprises a server farm having a plurality of servers and computer resources coupled to said first and second edge routers by a router and a load balancer.

13. The data center system of claim 12 wherein said servers use route health injection (RHI) to inject static host routes of interest to said router if said server farm is in an operational state.

14. The data center system of claim 13 wherein said router redistributes said static host route into its OSPF routing protocol and advertises said route to said first and second edge routers.

15. The data center system of claim 14 wherein said first and second edge routers use an IP Anycast mechanism to load balance client requests.

16. The data center system of claim 12 wherein said router is coupled to said first and second edge routers by an IGP link.

17. The data center system of claim 10 wherein each of said first and second edge routers is linked to a multilayer switch feature card.

18. The data center system of claim 11 wherein said one of said first and second edge routers configures a first metric for the route updates sent to a router at said first internet service provider and a second metric for the route updates sent a router at said second internet service provider such that said first internet service provider is configured to be the preferred path to said first and second data centers.

19. The data center system of claim 18 wherein inbound routes to said first and second data centers are set so that said first and second edge routers select said first internet service provider as a next-hop router.